VMware Communities : Document List - Enterprise Strategy & Planning

Upgrade path from acceleration kits?

Thu, 08 Oct 2009 13:59:42 GMT

Hello all. Faced with choosing a virtualization platform and my head is spinning trying to compare Hyper-V, Xenserver, and VMWare. Bottom line is I would go with VMWare in a heartbeat if cost were not an issue, so I am looking at the acceleration kits.

The problem is...what happens when you outgrow the vcenter 'f3 server limit? Is there any sort of upgrade path?

Take a look at the upgrade center for vSphere licensing: http://www.vmware.com/products/vsphere/upgrade-center/licensing.html

Here is a snippet regarding acceleration kits:

Acceleration Kits:

Licenses which were originally purchased as part of an acceleration kit will be treated as individual products (e.g., VMware Infrastructure 3.x Standard). The upgrade path for these individual products can be found in the table above, as noted for Edition Licenses. You'll find a table if you search for 'Acceleration Kits' on the page and the upgrade path

=========================================================================
William Lam
VMware vExpert 2009
VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/
vGhetto Script Repository
http://twitter.com/lamw

If you find this information useful, please award points for "correct" or "helpful".

Thanks William.

I should have been more clear. I currently do not own any acceleration kits. I mean that if I purchase a VSphere 4 acceleration kit today and find that I start needing to add and manage more ESX hosts down the road, what would the upgrade/licensing costs look like?

You will be able to upgrade but I am not sure what the price - I would contact VMware Sales or your Reseller and they will give you the price of moving form the Essentials version to the enterprise -

In regards to the competitors remember you get a higher density with VMware requiring less harware, cooling and power - also when comparing do not forget to include the cost of the management components -

If you find this or any other answer useful please consider awarding points by marking the answer correct or helpful

It is nowhere near that clear.

In VI3, accel kits were standrd VI# and the Foudation edition of VCentre. There was a clear upgrade path for vCentre.

The docs for vSphere do not show an upgrade path out of Essentials. Plus, the comparidon chart on this page http://www.vmware.com/products/vsphere/buy/editions_comparison.html seem to inducate that tere is a real differencr etween vCentre Essentials and vCentre Foundation/full.

VMware needs to come out with definitive statements regarding the upgrade path out of Essentials, as well as Essentials compatibility with other offerings like SRM.

As I understand it there is no difference in the code base between vSphere Essentials or the other vSphere products it is just a change in the license that you are applying - so to move up to the higher feature sets you will just need to pay the up charge - for the best information is to contact VMware sales or your local VMware reseller/VAC

If you find this or any other answer useful please consider awarding points by marking the answer correct or helpful

Thanks everyone. I did speak with an inside sales rep and from I understand, it would be a matter of going from Vcenter foundation to Standard...so maybe like around $3500+ , which would not be the end of the world for us, but not cump change for us, either.

Hard pill to swallow for an SMB.

This document was generated from the following thread: Upgrade path from acceleration kits?

VMFS vs RDM (vs VMDirectPath and other solutions)

Fri, 18 Sep 2009 15:15:15 GMT

VM disk type

There are different solution for implement a VM disk:

(native) vmdk over VMFS datastore (see also the different VMDK virtual disk type)
(native) vmdk over NFS datastore (vmdk format usual is thin)
virtual RDM (not for NFS datastore)
physical RDM (not for NFS datastore)
NPIV RDM (only for FC storage)
(native) guest iSCSI
direct with VMDirectPath I/0 (see also VMware VMDirectPath I/O)

How to choose virtual disk for a VM, iSCSI initiator in ESX, iSCSI initiator in OS or RDM

Performance difference

http://www.vmware.com/files/pdf/vmfs_rdm_perf.pdf
http://www.vmware.com/files/pdf/performance_char_vmfs_rdm.pdf
http://www.virtualization.info/2008/02/benchmarks-vmware-vmfs-vs-raw-disk.html
http://it20.info/blogs/main/archive/2007/06/17/25.aspx

Virtual Compatibility Mode Compared to Physical Compatibility Mode
http://pubs.vmware.com/vi35/server_config/wwhelp/wwhimpl/common/html/wwhelp.htm?context=server_config&file=sc_adv_storage.12.6.html

VMDK disk vs RDM disk

VMFS vs. Physical Disk performance
vmfs or raw for vms? which to choose
Question re RDM vs VMDK for high performance
RDM vs. VMFS...again...
vmdk vs RDM for large disk
VMFS vs raw data mappings in windows 2008.
Difference between a VMDK disk and Virtual RDM disk

Usually the "keep it simple" approach is the best choice... and vmdk over VMFS are very simple...

But anything that adds an "additional layer of abstraction" must therefore create a (no matter how small) overhead. For this reason when you have mid-range to high-range I/O on a disk, the RDM will perform a little better than a vmdk inside a VMFS.

For any database over a reasonable size I could be a good idea to create the database drive as an RDM - first to remove any I/O hit, and second to give the flexibility to access that RDM-ed LUN on a physical box without the need to 'convert' or V2P it.

Remeber one thing that a lot of people forget... the VMFS mounts from your SAN come to the hosts over the SAME fibre infra as the RDMs (same switch, same HBA etc) - if one VMFS-hosted VM is leech-whoring all the FC bandwidth then having your database be on an RDM is not going to help matters, is it?

VMDK vs VMDirectPath I/O

http://kb.vmware.com/kb/1010789 - Configuring VMDirectPath I/O pass-through devices on an ESX host
VMware VMDirectPath I/O

Actually VMDirectPath I/O it the best solution for VM with very high I/O, cause the performance are like in "native" mode.
But you loose a lot of the advantages of the virtualization: no VMotion, no backup, no cold migration between ESX, ...

VMDK vs native iSCSI

Re: iSCSI Performance
Creating VMFS on multiple internal SCSI disks

This solution could be very simple and "natural" in a iSCSI environment.
But remember that VM with a guest iSCSI cannot be protected with a backup solution for virtual environment, cause VCB, VDR or similar program can not "see" the iSCSI disks...

How to build a SAN

constantin.videnski@starwindsoftware.com — Fri, 28 Aug 2009 11:02:16 GMT

Storage Disaster Recovery

constantin.videnski@starwindsoftware.com — Thu, 13 Aug 2009 13:26:21 GMT

SQL Server Clustering Guide

constantin.videnski@starwindsoftware.com — Thu, 13 Aug 2009 07:46:02 GMT

The 2010 Guide to iSCSI Storage

constantin.videnski@starwindsoftware.com — Wed, 12 Aug 2009 14:55:15 GMT

Lab review: Cost-effective iSCSI SANity

constantin.videnski@starwindsoftware.com — Wed, 12 Aug 2009 14:29:01 GMT

How to increase VMware application with shared storage

constantin.videnski@starwindsoftware.com — Wed, 12 Aug 2009 08:12:25 GMT

Disk partition alignment

Sun, 09 Aug 2009 07:01:03 GMT

What is Disk partition alignment?

In a SAN environment, the smallest hardware unit used by a SAN storage array to build a LUN out of multiple physical disks is a called a chunk or a stripe. To optimize I/O, chunks are usually much larger than sectors. Thus a SCSI I/O request that intends to read a sector in reality reads one chunk.
On top of this, in a Windows environment NTFS is formatted in blocks ranging from 1MB to 8MB. The file system used by the guest operating system optimizes I/O by grouping sectors into clusters (allocation units).
Also, operating systems on x86 architectures create partitions with a master boot record (MBR) that consumes 63 sectors. This is due to legacy BIOS code from the PC that used cylinder, head,
and sector addressing instead of logical block addressing (LBA). Without LBA, the first track is reserved for the boot code, and the first partition starts at cylinder 0, head 1, and sector 1. This is LBA 63 and is therefore unaligned.
An unaligned partition results in a track crossing and an additional I/O, incurring a penalty on latency and throughput. The additional I/O (especially if small) can impact system resources
significantly on some host types. An aligned partitions ensures that the single I/O is serviced by a single device, eliminating the additional I/O and resulting in overall performance improvement.

The performance degradation of unaligned partition occurs during intensive I/O workloads rather than on those with low to moderate I/O activity.

A good document about align problem is "Recommendations for Aligning VMFS Partitions" (This paper is now considered obsolete, but it still contain a lot of useful information):
http://www.vmware.com/pdf/esx3_partition_align.pdf

For the latest information on this topic there "Performance Best Practices for VMware vSphere® 4.0":
http://www.vmware.com/pdf/Perf_Best_Practices_vSphere4.0.pdf

Align a VMFS partition

Note that if you add a datastore using a VI Client connected directly to the ESX server or through vCenter your disk will be aligned .
For manually align a VMFS partition, see:
http://www.vmware.com/pdf/esx3_partition_align.pdf

These recommendations are for block-based storage solutions. I/O characteristics on NFS are different from those of Fibre Channel and iSCSI storage systems. Though partition alignment eliminates track crossings and benefits performance on all storage platforms, the throughput improvements in specific types of I/O between SAN, NFS, and iSCSI are different.

Note that the first VMFS volume of a ESX 4.0 could not be aligned: the reason is that the volume contain the COS vmdk, so is not possible to free the volume to do the alignment.

Align a Guest OS vmdk disk

In Windows Vista as well as Windows Server 2008, partition alignment is usually performed by default. The default for disks larger than 4 GB is 1 MB
Partitions created on versions of Windows up to and including Windows Server 2003 by default are not aligned. Partition alignment must be explicitly performed.

For Microsoft OS see also:
http://support.microsoft.com/kb/929491
http://technet.microsoft.com/en-us/library/aa995867.aspx
http://download.microsoft.com/download/C/E/7/CE7DA506-CEDF-43DB-8179-D73DA13668C5/DiskPartitionAlignment.docx

More info

Storage Alignment and VMFS Block Sizes
Re: I/O performance of vSphere
VMFS Partition Alignment

Specific Software

To check if your VMs are aligned you can download a free tool called VM Check Alignment available at http://ctxadmtools.musumeci.com.ar/VMCheckAlign/Default.asp

To perform alignment there are some commercial product, like Vizioncore vOptimizer Pro:
http://www.vizioncore.com/products/vOptimizerPro/features.php

VI3 on IBM Bladecenter H - High-Level Storage--Datastore Provisioning

Wed, 15 Jul 2009 18:59:30 GMT

VI3 on IBM Bladecenter H - High-Level ESX

Wed, 15 Jul 2009 18:35:18 GMT

VI3 on IBM Bladecenter H - High-Level Blade Network Config

Wed, 15 Jul 2009 18:26:29 GMT

VI3 on IBM Bladecenter H - High-Level Chassis Network Config

Wed, 15 Jul 2009 18:14:23 GMT

ESX on Workstation 6 - Part3 OpenfilerVirtualSharedStorage

Mon, 23 Feb 2009 12:50:14 GMT

Securing VMware ESX servers

Sat, 21 Feb 2009 19:36:49 GMT

Microsoft Cluster Server (MCS) versus VMWare HA

Tue, 10 Feb 2009 18:03:21 GMT

Windows 2000 server bare metal restore from physical server to ESX 3.5 guest OS with Tivoli Storage Manager

Mon, 02 Feb 2009 18:34:42 GMT

Below is a step by step guide on how to perform a bare metal restore of a Windows 2000 server from a physical server to an ESX 3.5 guest OS.
Reasons for a crash are numerous; in this case it's a DELL PowerEdge 4600 with a motherboard integrated RAID card that's broken.

If this crash was on a database server and you don't have LOCKED or OFFLINE database backups you could be missing a good amount of data.

I have personally used this exact process to restore Windows 2000 Servers. This specific documented repair was performed without having domain administrative rights, local administrator rights, a "working" ERD or a TSM performed system object restore on the system. If you don't take snapshots this can be a very long restore process for you if you have to start over a few times. I would recommend you build this system on a LUN \ VMFS that has at least 10Gb of free space for snapshots (dep
I provide no guarantee this will work for everyone in every situation.
Use caution and common sense when performing system restores!
*Basically...Try this at your own risk!*
Read it completely before starting any of the below steps.

Step 1;
Configure a new VMware guest
Storage for the C:\ should be about 2.25 times the original size of the DATA that was originally on the drive.
(You need space for the restore and additional files and growth. Remember the 15% requirement for defrag's, etc)
Match or best the storage of any additional drives by 10% to 20%. (This will be directly related to how much free space you have and how much growth you're planning.)

Step 2;
Install Windows 2000 server
Use the same server name of the crashed system (and IP's if you want).
Install the recovery console; access the command window (start | run | cmd).
connect your OS disk and type; e:\i386\winnt32.exe /cmdcons
(e:\ can be replaced with whatever drive is your CD/DVD rom)
Reboot and ensure you can log into your Recovery Console partition.
Do not continue if you can't use the recovery console properly, it (or an equivalent) will be required later during the restore.

Step 3;
Install Tivoli Storage Manager (TSM) preferably version 5.4 or higher.
Check your documentation or patch manager (WSUS) for the correct patch level (SP2, 3, 4, etc).
Log into TSM from the new system and review the installed windows updates.
In TSM select restore and navigate to;
File level | servername\C$ | WINNT
Inspect each patch's $NT or $MS or any other $ update directory for updates that modify the following files;
hal.dll, kernel32.dll, ntdll.dll, ntoskrnl.exe, ntkrnlpa.exe, win32k.sys, winsrv.dll and comsvcs.dll
Note each of the patch KB# names

Step 4;
Match the ServicePack level of the original system.
Then download and install all of the patches and service pack level you noted above in "Step 3" rebooting after each install.

Step 5;
Create a directory named reg-repair under C:\WINNT\
Create a directory named new-reg under C:\WINNT\reg-repair\
Create a directory named old-reg under C:\WINNT\reg-repair\
Create a directory named drivers under C:\WINNT\reg-repair\
Create a directory named boot under C:\WINNT\reg-repair\
Create a directory named sys32 under C:\WINNT\reg-repair\
Create a directory named reg-exp under C:\WINNT\reg-repair\

Step 6;
Copy from C:\WINNT\REPAIR the SAM, Security, Software and System registry hives to C:\winnt\reg-repair\new-reg
Copy from C:\ BOOT.ini, NTLDR, config.sys, io.sys and NTDETECT.COM to C:\WINNT\reg-repair\boot
Copy from C:\WINNT\System32 hal.dll, kernel32.dll, ntdll.dll, ntoskrnl.exe, ntkrnlpa.exe, win32k.sys, winsrv.dll and comsvcs.dll to C:\WINNT\reg-repair\sys32
Copy from C:\WINNT\System32\Drivers DISK.SYS, SCSIPORT.SYS, vmmouse.sys, vmscsi.sys, vmx_svga.sys and vmxnet.sys to C:\WINNT\reg-repair\drivers

You may need to enable "Show Hidden files and folders" and uncheck "Hide protected operating system files" from;
Open C:\ , select "tools" then folder options, then "view" tab and check/uncheck the appropriate box and click apply.

Step 7;
If your OS and DATA drives are the same drive IGNORE THIS STEP!!!
If you have a second drive for data or applications format that disk drive and log into TSM and perform a restore of that data drive.

Step 8;
Take a snapshot and distinctly label the snapshot something like;
Working base OS, fully patched and system files copied.

Step 9;
DO NOT REBOOT
Open the TSM application and select restore.
Check your OS volume as the restore item.
While in the restore window select;
Options (button)
All selected files and directories (check Restore NTFS security information)
Set; Action for files that already exist to; "Replace"
Check Replace files even if read-only/locked
Select restore, and select restore in original location.
Ignore any windows that warn about rebooting.
Again, don't reboot!

Step 10;
Now that your C:\ is restored completely take another snapshot. Label this snapshot with a good description.
Something like System-x complete data and OS restore, system not operational.
Don't reboot.

Step 11;
Copy from C:\WINNT\reg-repair\boot BOOT.ini, NTLDR and NTDETECT.COM to C:\ replacing the ones just restored.
Copy from C:\WINNT\reg-repair\sys32 hal.dll, ntoskrnl.exe, ntkrnlpa.exe, to C:\WINNT\SYSTEM32 replacing the ones just restored.
The other System32 files are backups/just incase files, they are currently in use anyway and you'll receive a copy error.

Step 12;
Navigate to C:\WINNT\repair\regback
You should see the old registry files from the original system. The ones replaced in C:\WINNT\Repair may not be the correct date.
The RegBack directlry will not likely exist if you haven't previously created an ERD and backed up the registry, or in general backed up the registry.
Copy from C:\WINNT\REPAIR\regback the SAM, Security, Software and System registry hives to C:\winnt\reg-repair\old-reg

Step 13;
Reset your Administrator password to whatever you want, just reset it even if it's the same password.
Take a snapshot and describe it as something like; System restored, boot and system32 files replaced - pre-recovery console operations.

Step 14;
Reboot
At the boot option window press F8 to enter Directory Services Restore mode. Regardless if this is a DC or not.
Enter the recovery console.
Log into your recovery console OS partition

Copy from C:\WINNT\Reg-repair\Drivers DISK.SYS, SCSIPORT.SYS, vmmouse.sys, vmscsi.sys, vmx_svga.sys and vmxnet.sys to C:\WINNT\DRIVERS\
Copy from C:\WINNT\Red-Repair\New-Reg\ the SAM, Security, and System registry hives to C:\WINNT\CONFIG
Copy from C:\WINNT\Red-Repair\old-Reg\ the Software registry hive to C:\WINNT\CONFIG
Copy from C:\WINNT\reg-repair\sys32 hal.dll to C:\WINNT\SYSTEM32 (should already match, it's a just incase step).

exit the recovery console and reboot

Step 15;
At the boot option window press F8 to enter Directory Services Restore mode. Regardless if this is a DC or not.
Choose your Windows 2000 Server install as your boot option.

You should not receive a 007B BSOD disk error or a NTOSKRNL error, if you do;
You may also need to copy the additional files from Boot and System32 that were not required in the previous recovery console operation.
If so, try io.sys and config.sys first, the others from System32 may potentially down-level your OS build to the disk installed version.
Also review the steps and ensure you didn't miss anything.

If you didn't get the BSOD or NTOSKRNL then....
Boot into Windows and login.
Let the system install any Plug-and-Play devices.
Take a snapshot and label it something like; System restore process after recovery console repair tasks.

Step 16;
Install a registry editing program to review this file; C:\WINNT\reg-repair\old-reg\system
Something freeware, or something like "Alien Registry Viewer" works fine. This step is for comparison only.
Specifically under HKLM\System\CurrentControlSet00#\services
(Replace the # above with each specific CurrentControlSet key you have.)
Navigate to the running registry with regedit in windows and view the current registry settings. ( Start | Run | regedit )

Step 17;
With both SYSTEM registry components open compare the ones from the C:\WINNT\REG-REPAIR\OLD-reg\system to the current settings in regedit.
Under HKLM\System\CurrentControlSet00#\services you will need to find the service you're missing for the critical application, and add the keys in to the regedit window manually.
Once you've manually edited the keys for the first CurrentControlSet00# export each key that you edited to: C:\WINNT\Reg-Repair\REG-EXP\
Right click each exported registry key and modify the CurrentControlSet00# for each # you need to import. Re-import each key. This prevents you from having to manually re-create each registry key.

Step 17;
Take a snapshot and describe it as; System is operational and services have been created and imported.
Reboot the server
At the boot screen select the Windows 2000 Server operating system and boot normally.

Step 18;
Check your services everything should be in place and operational.

Reinstall TSM and any other programs you didn't feel like resetting the registry service keys for, or were not part of the older software hive.
Reinstall the VMware tools.
Check your scheduled tasks and set the passwords appropriately.

Step 19;
Have your Domain Administrators delete the previously existing domain entity for the crashed server.
Re-Add this new server to the domain and the appropriate groups/OU's.

Add users or service accounts if needed.
reboot

Step 20;
Review your documentation for previously configured network shares.
If you don't have any documentation open up C:\WINNT\REG-REPAIR\OLD-REG\system with that program from "Step 16".
Review this key; HKLM\System\CurrentControlSet00#\services\lanmanserver\shares

Your old shares are in that key.

Re-enable those shares and add the appropriate users.
You should not have to restore the NTFS security permissions; they should still be intact (from Step 9).
The users GUID's will resolve to names once the server is successfully part of the domain once again (previous Step 19).

Step 21;
Take a snapshot for the current working state.
Delete all other old snapshots with the exception of the latest working one.
Backup the previous C:\WINNT\Repair\RegBack and take a fresh ERD and add it to your ERD repository.
Double check TSM scheduled backup processes.

Wait for your backup process to run a few cycles before you delete any remaining snapshots for this system.

Step 22;
Work on one of the other 1002 things you have pending.
Make that 1003, because you're going to plan an upgrade of this system to Windows 2003 to use ASR for bare metal restores.

Goodluck with your restore!

VI to VI migration - Howto - Best Practice with minimal downtime

jo.strasser@salk.at — Wed, 28 Jan 2009 01:02:32 GMT

ALL THIS INFORMATION WASN´T FIXED TO HARDWARE-VENDORS.
U ONLY NED A VI3 INFRASTRUCTURE OF VMWARE (tested with ESX3.5 - vCenter2.5 - HP EVA8k and NetApp FAS31xx Storage Systems over FC).

My name is Johannes Strasser, working in a primary HealthCare Company in Austria (SALK Ges.m.b.H.) and I want to describe a fast step-by-step migration of a complete Virtual Infrastructure (VI) with minimal downtime over Fibre Channel.

Our Company is a 7/24, we haven´t many time for maintenance.
So I´ve write a how-to for the fastest migration of VMs: over FC-Network.

History:

We have at this moment over 310 VMs, they should all be migrated to a new build VI... new Hardware for Hosts, vCenter, and new Storage Systems,... with a minimal of downtime.

Many Users make a mistake, they move VMs over a "standard-offline-migration".
They shutdown VMs, and migrate them with the command "migrate" over vCenter.

So, the VM moves, but over the network, with low performance.
Now, I will describe an alternative to migrate VMs, over FC:

Before U migrate:

Present one host of your new "VI" all old and new LUNs of all Storage Systems (source-destination).

Best Practice for Migration over FC-Network:

1.) Add this host to the old vCenter (if U have)
2.) Migrate VMs to this Host "offline" without Disk-Migration (keep disks)
3.) Remove Host from this DataCenter/Cluster (keep evacuating VMs, if U want to enter Maintanence Mode)
4.) Add this Host to the new Datacenter/Cluster
5.) Power on all moved VMs, all this where online and working again on the same ESX-System
6.) Check Storage Availability (DiskSpace) on Destination - Storage System
7.) Start "Storage Migration" (it will give addons for vCenter!) - or over Windows PowerShell (scripts)
8.) If U finished all Storage-VMOs, online-migrate (vMotion) VMs to other Cluster Hosts, or set the "Miration-Host" in maintenance mode to evacuate all VMs to the right position,... to the original new VI...

U can repeat this steps to move more and more VMs...

In 2 hours, I think, U can move about 20-40 VMs without dramatic downtime.

Thanks.

For Feedback, please give it over the VMTN (VMware Technology Network) Community.

Strasser Johannes, VCP, CPSA
Systems Engineer / Virtualization Specialist
________________________________________________________________
Gemeinnützige Salzburger Landeskliniken Betriebsgesellschaft mbH,
Servicebereich Informatik und Medizintechnik,
Müllner Hauptstrasse 48, 5020 Salzburg, AUSTRIA

Tel.: +43(0)662-4482-4950
Mob.: +43(0)664-604141-4950
Fax +43(0)662-4482-4908

mailto:Jo.Strasser@salk.at
Internet: http://www.salk.at

A multi storage vendor post to help our mutual iSCSI customers

v@netapp.com — Mon, 26 Jan 2009 16:24:23 GMT

A very nice article was published by EMC, NetApp, Dell, HP, & VMware.

It can be found at the following locations:

http://blogs.netapp.com/virtualstorageguy/2009/01/a-multivendor-p.html

http://virtualgeek.typepad.com/virtual_geek/2009/01/a-multivendor-post-to-help-our-mutual-iscsi-customers-using-vmware.html

Creating task based roles in VMWare

Fri, 09 Jan 2009 15:15:11 GMT

Best Practices for Virtual SAP Infrastructure

Fri, 09 Jan 2009 14:50:53 GMT

Should DC, AD, DNS, DHCP servers be in VMware env?

Tue, 21 Oct 2008 14:13:46 GMT

Hi everyone,

Although recovery time for PS (Physical Server) is ridiculously long compared to that of VM, we currently cannot convince people to virtualize these types of servers. The argument was the fact that if something goes wrong -- long after the servers are virtualized -- we must reproduce the problem in PS before the vendor would even consider troubleshooting it.

Q1: Have anyone out there had ESX hosted these types of servers? What are the keys to make this happened?

Q2: Generally, if someone says "my vendor said that application is not supported in VMware" what should the approach be?

Q3: Is there a tool to convert back to PS ... like V2P?

Thanks!

PS: If this is the wrong discussion group, please point out the proper one.

A1:We host an old NT4 PDC in a VM - it actually performs much better. Basically it went there because the physical server was old and failing.

Many people run those servers in VMs, depends on the size of environment in my mind.

A2: DC, AD, DNS, DHCP are all microsoft (Im assuming) and if you have premier support they will support it in a virtual environment. Other apps its pretty much SoL, Ive ran into a few vendors that say that and its mostly just laziness on thier part to test thier apps.

A3: there are a few tools that do that, powerconvert from platespin does, HP also has an offering.

Every vendor has a different VMware support policy. Almost all vendors support it in some fashion. Microsoft is stubborn because they have a competing virtualiztion product. They will make best effort to support there product on Vmware and only ask to reproduce the problem on a physical system if they suspect it is related to the VMware enviroment (which it usually isn't). If you have premier support with them I beleive they will not ask you to do this. I have half my DC's virtualized, that way if there is a domain problem I have physical one's also. In the planning phase of my virtualization project I researched Vmware support statements from each vendor that was going to be virtualized. If the vendor says they do not support Vmware, ask them why. Alot of times it's because they have never taken the time to try their product on Vmware. In 99% of the cases it will run just fine on Vmware. You can test it and see, put your dev/tests systems on Vmware and if they are fine after several months then migrate your production systems.

Virtual Machine to Physical Machine - Migrationhttp://www.vmware.com/support/v2p/doc/V2P_TechNote.pdf
Microsoft Virtual Machine Technology FAQ - http://www.microsoft.com/licensing/highlights/virtualization/faq.mspx
Virutalization of Active Directory - http://www.vmware.com/community/thread.jspa?messageID=352424&#352424
Considerations when hosting Active Directory domain controller in virtual hosting environments - http://support.microsoft.com/kb/888794
Virtualizing a Windows Active Directory Domain Infrastructure - http://download3.vmware.com/vmworld/2006/tac9710.pdf

fyi...if you find this post helpful, please award points using the Helpful/Correct buttons...thanks

I would never put your FSMO and primary DFS and GPO host in Virtual, just due to pure performance and instantaneous needs, but any secondary / periphery systems you might have should be fine as VM's, no special circumstances _need_ to be considered as far as I can tell. None of them do anything funky with the network, all normal TCP/IP traffic.

Be aware of the growth of the DS in terms of disk size so you don't have to worry about taking DC down to expand the disk, etc.

It might be a good idea to put reservations on the resources for the VM's, just so their instantaneous needs can be better met.

If a vendor says they won't support the application in VM, just make them aware that if they don't, their product will fall by the wayside as others fill that market - but say it in the nicest way possible

Why would you never put FSMO roles on VM'S? Schema Master, Domain Naming Master, RID Master and Infrastructure Master are not that resource intensive or used on a constant basis. PDC Emulator is the only role that I could see being utilized frequently. You can also make multiple DC's Global Catalog servers to serve as backups to each other.

Yah sorry, I started thinking our environment and our "DC01" is the big honco, PDC Emulator, everything etc.

Q1: Yes / Why to you run other servers as VMs? [Put your reason here]
Q2: Some vendors are not (yet) familiar with vitualization. Try to show them the benefits and do a pilot with them to show them that it simply works. Try to get them into your boat. That usually works. - If that doesn't help it's always good to inform that vendor the other concurrence vendor supports virtualization ...
Q3: PlateSpin and others do that. Here's the official VMware point: http://www.vmware.com/support/v2p/

As said in this thread already all of these microsoft services virtualise with very little difficulties .

IF you're having a problem selling it, for me you need only look at the advantages that Vi3 provides;

HA: High Availability, abstraction from hardware failures
DRS: Resource Scheduling to ensure VM performance across the cluster
VCB: LAN free backups...

What type of issues do you get with the physical servers, would virtulisation be increasing these or adding value to the service delivery.

Q1: Have anyone out there had ESX hosted these types
of servers? What are the keys to make this happened?Yep, we host pretty much entire back office infrastructure on ESX today. That includes *ALL* domain services, DHCP, network management suite, backup infrastructure, all exchange services except mailbox servers (bridgeheads, antispam, BES, etc), SQL, Oracle, etc... I could go on The key to making it happen in our case was buy-in from the top of our organization that this was the right approach. Once that happened, everybody else sort of fell in line..

Q2: Generally, if someone says "my vendor said that
application is not supported in VMware" what should
the approach be?Generally, the way I see it, is that a software vendor should be hardware agnostic. With that said, many still will play the card of "we'll provide commercially viable means to help troubleshoot, buuut.. we may ask you to convert back to a physical server". I'm actually in the middle of a project where I keep hearing that, and I keep asking the same question of these vendors (I phrase it one of the following two ways):

- "Other than being potentially faulty, something that is minimized when talking about virtual machines, what does the type of hardware that your application runs on have anything to do with if it's functional?"

- "What does the hardware have to do with anything? Aren't we talking about software here? Shouldn't the operating system, and it's various patch levels, have more to do with issues?"

I see this "lack of support for virtualization" to be a ploy by software manufacturers to not have to support a new platform, but the irony of it is that in a lot of cases, these products are being developed and tested internally on some virtualized platform... It's funny how things work, huh?

Hi everyone,

Although recovery time for PS (Physical Server) is
ridiculously long compared to that of VM, we
currently cannot convince people to virtualize these
types of servers. The argument was the fact that if
something goes wrong -- long after the servers are
virtualized -- we must reproduce the problem in PS
before the vendor would even consider troubleshooting
it.

Q1: Have anyone out there had ESX hosted these types
of servers? What are the keys to make this happened?A LOT of people have their DCs virtualized. DNS and DHCP is even more trivially virtualized.

Q2: Generally, if someone says "my vendor said that
application is not supported in VMware" what should
the approach be?It depends on the application--there are things that don't virtualize very well, and the vendor may have a legitimate reason (like ESX doesn't support hardware that their application talks to etc.)

Then again the vendor may be back in the horse and buggy days, in which case you should assist the market in selecting them out.

Seriously. Other than applications that need SERIOUS hight performance, or are oddly hardware dependent you have to have written your product fairly badly not to work inside a VM.

(For example it doesn't make a lot of sense to deploy a Bewoulf Cluster on ESX, nor would it make a lot of sense to deploy a Real Time OS that controls CNC machinery etc.)

Q3: Is there a tool to convert back to PS ... like
V2P?You could do it with Ghost or Alteris like tools, if you have the right drivers already installed or available.

A couple other thoughts.

Often people have these services running on old servers, 3-5+ year old servers, and to me the risk of staying on these old servers is higher than moving to a VM. Old servers are more likely to have hardware failures.

Since there is more than one (DC, DNS, DHCP, etc) server, start by virtualizing one of each. We have had several DCs, secondary DNS, DHCP, and WINS all running in VMs for a year now. Everyone is now comfortable with moving the rest of this infrastructure to VMs.

Most first level support you deal with wont ask, or even know your application is running in a VM, don't offer that up either IMO. Also, reproducing on physical hardware should be the last step, after all other possibilities are examined. My personal favorite P2V and V2P tool would be PlateSpin PowerConvert, with this you can basically go any direction.

Most of ISVs says that they not certificate their applications if they run on VMs, but the fact is that the applications talk only with the OS and not with the Hardware (HW). VMware Virtualizes the HW, but for applications it really don't matter. Then, if your OS is certified for VMware you should have no problems for support.

Now, talking about your question, this services may run on VMs without problems, except DNS, because ESX and VirtualCenter use DNS for name resolution. If you want, you can configure hosts file for name resolution to workaround this problem, but if you use VMware HA, you need DNS. Then you can configure a DNS on a physical server and leave all your other services (include alternate DNSs) in VMs

SLT

We host multipes of these for several companies.
12 DC's for 6 different domains 10 ar AD
DNS for 2 companies
DHCP services for 3 companies.

As far as AD/DC how many users and queries would be the concern
These servers support 600 people total no issues, including vpn, and pt to pt tunnels

Q2 do not tell them it is a vm, until the request logs and see for themselves what the hardware is. Most
of vendors do not recommend specific hardware.

Q3 I ha succesfully used Bart PE to V2P several systems and 1 stayed as physcial due to the high utilization

Q1: Yes / Why to you run other servers as VMs? [Put
your reason here]

Oh I have tons of reasons. What I dont have is a selling skill... I'm tiny technical guy and MS reps are pretty big and heavy ;-(

Q2: Some vendors are not (yet) familiar with
vitualization. Try to show them the benefits and do a
pilot with them to show them that it simply works.
Try to get them into your boat. That usually works. -
If that doesn't help it's always good to inform that
vendor the other concurrence vendor supports
virtualization ...
Q3: PlateSpin and others do that. Here's the official
VMware point: http://www.vmware.com/support/v2p/-----

I am running DHCP, WINS and Radius on virtual servers running Windows 2003 on ESX 3.5 with VC 2.5. Not a single problem what so ever. Both of the dhcp servers were cloned using VMware convertor enterprise (Cold Clone). I also resized the hard disk since it was a waste to store 75GB for my dhcp server on my precious SAN storage .

Hi,

We are in the process of virtulizaing our entier MS Infrastructure from P.S to Virtual enviroment in upgrading the exisiting exchange 5.5 to exchange 2007. both the esx hosts will be in cluster mode and accessing HP MSA 1000 SAN in active/active. IN DR-Site, there will be the same setup with DL380 in cluster mode and also there will be an HP MSA 1000 in active/active mode, and the SAN merroring will be replicated from Primary Site to the DR Site.

This document was generated from the following thread: Should DC, AD, DNS, DHCP servers be in VMware env?

Small branch office server RAID setup

Wed, 10 Sep 2008 08:52:48 GMT

Hi,

We have 4 remote offices (from 15 to 40 users each) and are planning on putting a VMware setup in for each site.

Were going to go for an IBM x3650 (2x quad cores) with 6GB RAM and 6x 146GB 15k SAS disks and VMware ESX Starter for each site

Each of these machines will be running the 3 (or 4) VMs:
File, Print & AD
Exchange
Misc (voice mail, etc)

In the future we may want to put Terminal Server or Citrix VMs on these boxes and replicate these VMs to our head office for DR.

Does this seem a reasonable solution?

Would one of these machines handle all these services and potentially TS or Citrix? I realise it would be one machine running everything, that is why we would consider replication.

I have been recommended to setup 2x disk for ESX OS and 4 for data (RAID 5 or 10). However I think it would be better to setup a single RAID 5 with 1 hot spare. My theory is that the ESX would hardly be using the disks so it would be better for all round performance to have more spindles for the VMs. How would you setup the disks for best performance and storage space?

Thanks,
Roger

Would one of these machines handle all these services
and potentially TS or Citrix? I realise it would be one
machine running everything, that is why we would
consider replication. Yeah, it should be OK. Provided your Exchange server isn't too busy. Voicemail may be a problem, because that's fairly time-critical as far as voice processing goes.> I have been recommended to setup 2x disk for ESX OS
and 4 for data (RAID 5 or 10). However I think it
would be better to setup a single RAID 5 with 1 hot
spare. My theory is that the ESX would hardly be
using the disks so it would be better for all round
performance to have more spindles for the VMs. How
would you setup the disks for best performance and
storage space?Theoretically, you get better RAID5 performance with either 5 or 9 disks. 4 or 8 data stripes plus parity)More spindles is usually better, but if you have a drive failure, it becomes much slower because the controller has to read EVERY disk to reconstruct the missing data.

Depending how much space you need, you might find it more efficient to have more spindles but smaller drives. Smaller drives are generally more reliable, because the have fewer platters and less sector density.

Thanks for your help.

Were planning on having 6 disks at this stage, we can add other drives externally at a later stage if we need them.

Which option would be better?

Option 1:
2x disks for VI3 (RAID1)
3x disks for VMs (RAID5)
1x disk for hot spare

Option 2:
5x disks for VI3 and VMs (RAID5)
1x disk for hot spare

I'm thinking option 2 sounds better as we get more spindles. However we would no longer have dedicated drives for the OS (VI3). My thoughts are that it won't be using the disks much anyway so the VMs could do with the extra spindles.

Thanks,
Roger

I'de go with a Netapp S500 (1-6TB) for CIFS and use NFS on the S500 for your vmware datastores. Snapmirror all the CIFS and vmware volumes back to the head office (fas2020) for a tapeless backup/DR solution . Simple and expandable.

when you say 2 disks for ESX OS i presume you mean 2 x 146Gb mirrored? That is overkill as the ESX OS will use no where near any of that amount of storage. As pointed out, if you have the option go for smaller disks and more of them. You could then have 3 separate arrays

mirrored OS,

RAID5 for VM's,

RAID5 for exchange data

Pez

We have essentially the same setup, but with HP hardware.

We went with the 2x146GB disks for ESX and 4x146GB disks (RAID 5) for the VMs. The RAID 1 partition allows us to segment the setup a little bit, do backups to RAID1, etc. The RAID controller in the server has 512MB of cache, and does about 60-80MB/sec on the RAID5 volume, plenty fast for our remote needs.

I have struggled with a similar situation in my env. We use HP hardware so I went with a mirrored (raid 1) OS of 72 Gb drives (overkill I know but it is what came with the server) and then a Raid 5 or the ADG (you can lose 2 drives) data array to store my vmx and vmdk files. One thing to keep in mind is that this type of senario only protects you from disk failure. I actually had a server die on me a month after installation. It brought the whole office down. Thank fullly i had 24x7 support with 4 hour response time and we were able to get the box back up and running in short time but there are other issue to keep in mind when spec'ing a single server to handle an entire site. I have since moved some of my larger remote offices to a citrix solution out of my main datacenter to try to mitigate downtime. Now I am dependent on my WAN connection

the thing with ADG is it sucks up disc space and when that's at a premium you're stuck

think about it for a minute, the OS doesnt actually store much does it.

just use 1 x 72GB disc for the OS (you can always recover/reinstall from the cd in about 10-15 mins)

raid5 for VM's

raid5 for exch data

it really goes back to the old age adage "there's more than 1 way to skin a cat"

Thanks everyone for your help.

We are still considering centralising our system or not so this may or may not be a temporary solution. These are small sites of 15 to 35 users so putting in a SAN at each site is out of the question due to the cost. Since we may end up pulling these machines back into our data center we need to stick with drives we can re-use that's why we are going for the 3.5" SAS internal drives as we could then put them in our main site SAN (DS3400) if we centralise.

For now we have access to 6 drives max and are unlikely to get funding for any more. The current servers at each site are actually using 2x RAID1 arrays with 4x ATA drives and a P4 of some kind. I suspect that this solution should be heaps faster.

I am reluctant to use 1 disk for ESX as we have no onsite IT people and rebuilding would be a real pain, having redundancy is critical. A hot spare drive isn't really necessary but would be nice, I guess we could always have one sitting next to the server and get someone to swap it if it fails (assuming IBM can't replace it quickly).

Since we are limited to 6 disks then having multiple RAID5 for the VMs and Exchange are out unless we do 2x 3 disk RAID5.

Perhaps the best question is why should we NOT put the VMs and ESX on the same RAID array? Doing so gives us more options.

There are likely to be 3 VMs. File/Print/DC, Exchange & Misc.Load on these isn't that high.

Options we have that give enough disk space are:

2 disk RAID1, 3 disk RAID5, 1 disk HS (50GB spare)
2 disk RAID1, 4 disk RAID5 (190GB spare)
2 disk RAID1, 4 disk RAID10 (50GB spare)
3 disk RAID5, 3 disk RAID5 (150GB & 160GB spare)
5 disk RAID5, 1 disk HS (330GB spare)
6 disk RAID5 (470GB spare)
6 disk RAID10 (190GB spare)

I think option 5 is the best overall, my second would be option 2. If anyone can give me the perfect answer that would be nice.

Hi Everyone. I was just looking at ESX 3i. I assume when they are talking about this it is some sort of flash device inside the server or via USB flash disk (http://blogs.vmware.com/vmtn/2007/09/more-on-esx-ser.html)

If this is the case then I assume the hypervisor doesn't use much disk activity (if any). If this assumption is correct they why not have ESX and the VMs on the same RAID array for small setups?

Thanks,

Roger

3i is still Beta, but yes, it essentially boots the hypervisor off of a flash based device (USB key, flash disk, etc). No linux OS to manage/patch, etc. It would be ideal for this type of environment, but the only drawback is that it is beta, so if support is essential for you, it isn't the best solution out there. With that said, it is the future.

To answer your question regarding which RAID setup is best, the answer is: "it depends" (flashes of VMWorld VCB anyone?). The seperate RAID 1 volume is nice to store templates on, etc, etc. It is also a nice spot to run backups/snaps to without impacting your "production spindles".

If it were me I would try and allocate as much ram as possible to your DC to get it off of your disk. Try allocating enough memory to the DC to store your AD database in memory.

For the print server, definitely assign a second drive for the print queues, and you can play around with putting this on RAID5 or RAID1. The benefit to having these two arrays is that your are segmenting your spindles. In our setup the controller is very powerful and runs at about 1-5% CPU during peak hours, so can tolerate a split array type setup with no issues (512MB of cache is nice).

In this situation you just need to test, test, test.....It really depends on what you want out of the environment (i.e. ideal redundancy, performance, etc), and that will really drive your decision.

Good luck

i prefer option 4 as you definitely want to keep your exchange databases on a separate array

Thanks everyone. I still don't know what to do but will work that out. I appreciate all your help.

This document was generated from the following thread: Small branch office server RAID setup

iSCSI crossover and VI3

Wed, 27 Aug 2008 01:17:42 GMT

I have a single ESX in my DMZ and we are replacing it with a HA cluster of two 3.5x ESX Enterprise ed. These two ESX talk to our 2.5x VC server.

Question: we are considering purchasing a Dell MD3000i SAN that the two ESX will share using iSCSI. The MD3000i has four ethernet ports for connecting hosts. We want to connect the two ESX to the four ports using crossover cables on a dedicated iSCSI subnet so there is no switch involved.

Here are some facts:

The VC server will not be able to communicate with the SAN.
The two ESX will not be able to communicate with each other on the SAN subnet but they can communicate with each other on another VMKernel interface which has a SC.

Example of networking:
ESX1:
-SC/VMotion-enabled addresses 10.10.10.5, 10.10.10.6
-SC/iSCSI addresses 172.16.0.5, 172.16.0.6
-VM port group addresses 10.10.10.x

ESX2:
-SC/VMotion-enabled addresses 10.10.10.7, 10.10.10.8
-SC/iSCSI addresses 172.16.0.7, 172.16.0.8
-VM port group addresses 10.10.10.x

MD3000i SAN:
-SC/iSCSI address 172.16.0.4

Will this work? I'm worried it won't because the SC on the data subnet (172.16.0.x) won't be able to communicate with the other ESX SC.

Thanks!!

Scott

If you plan to use the software iSCSI initiator then this will not work - the SC must be able to communicate to the iSCSI SAN - easiest way to correct would be to add an SC port vSwitches you are using for the iSCSI VMkernel ports -

Hello,

Cross Over Cables will not work unless you also add a SC portgroup to the vSwitch to which you added the crossover cables.... This form of Direct Connect while possible is not supported by VMware. Also, even if you use iSCSI-HBAs the SC must participate in this network for authentication purposes even if you are not using CHAP. I would urge you to consider even a small gigE switch for this instead of cross over cables.

Best regards,
Edward L. Haletky
VMware Communities User Moderator
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education.
CIO Virtualization Blog: http://www.cio.com/blog/index/topic/168354
As well as the Virtualization Wiki at http://www.astroarch.com/wiki/index.php/Virtualization

Thank you for your replies. I'm having a bit of trouble getting my head around this scenario... my internal VI3 networking is fully available to our VC server but in this DMZ scenario we want to restrict traffic as much as possible.

Would you please look at this:

1. vswitch#1: vMotion subnet - VLAN off so only the ESX hosts can communicate? Does the VC server need to communicate on this subnet? Is it necessary for a SC on this subnet?

2. vswitch#2: VMKernel/ESX software iSCSI/SC subnet - VLAN off so only the ESX hosts and the SAN can communicate? Does the VC server need to communicate on this subnet?

3. vswitch#3: SC - VC server must communicate.

4. vswitch#4: VM port group - VC server must communicate.

Thanks again,

Scott

One comment here - for Gbit don't use the crossover - only cat 5e patch or higher.

Thanks Christian, I forgot to mention I would use a switch for the networking rather than a crossover.

Scott

SBaldridge wrote:
Thank you for your replies. I'm having a bit of trouble getting my head around this scenario... my internal VI3 networking is fully available to our VC server but in this DMZ scenario we want to restrict traffic as much as possible.

Would you please look at this:

1. vswitch#1: vMotion subnet - VLAN off so only the ESX hosts can communicate?

Yes.

Does the VC server need to communicate on this subnet? Is it necessary for a SC on this subnet?

No. The VMotion network can be totally isolated. In fact, for a two-host environment, you can use a direct cable connection between the two hosts to support VMotion.

2. vswitch#2: VMKernel/ESX software iSCSI/SC subnet - VLAN off so only the ESX hosts and the SAN can communicate?

Yes

Does the VC server need to communicate on this subnet?

3. vswitch#3: SC - VC server must communicate.

Yes

4. vswitch#4: VM port group - VC server must communicate.

Why? VirtualCenter has no need to communicate directly with the virtual machines. The VM remote console communication is handled across the service console interface. You should not have your VC server on a vSwitch that is visible from the DMZ.

Ken Cline
Technical Director, Virtualization
Wells Landers
VMware Communities User Moderator

Thanks everyone, I appreciate it.

Scott

This document was generated from the following thread: iSCSI crossover and VI3

IO LOCAL STORAGE ESX All Local Storage environment - performance degradation?

Fri, 15 Aug 2008 17:43:00 GMT

Greetings

I am assisting a group within my company troubleshoot some performance issues they are experiencing on a 2 host cluster. They are not using shared storage, rather a failover/load balancing solution incorporating a crossover cable between the two hosts. I think for what they are trying to accomplish the features lost in shared storage are somewhat understandable, however this is my concern:

They are running ESX and their Virtual Machine storage across 8 x 146gb drives in a RAID5. Some of the applications they have may be somewhat cpu and disk intensive. I am in the process of comparing their disk I/O statistics with other environments I administer, however since this would be a "new" way of configuring VM storage for me, I was curious of what input may be out there.

Currently they are running averages of 1,225 KBps and peaks (granted sharp and quick) as high as 71,000 KBps. I am not familiar enough with storage numbers to assess these as alarming or acceptable.

Any feedback is greatly appreciated!

Thanks

~grc

When you say local, I'll assume you mean DAS (Direct Attached). Storage is storage...DAS is not slower or faster than SAN. It all depends on the paths to the storage, the underlying subsystem, controller efficiency,hard disk type, number of spindles, etc.

For instance, I had a client running a DL580 G5 using (I think they have two P900 controllers with 10 drives each) direct attached SAS drives. They moved their VMs to an HP MSA2012 using 6 SATA drives in a single array (connected via FCP). They experienced WORSE disk performance after doing so. The problem was that they were basically going from a 10 spindle array of 15K SAS drives to a 6 spindle array of slower 7200RPM SATA drives. Also, FCP sounds nice and fancy, but wire speed is wire speed...data will travel faster from a local controller than a FCP network that also incorporates a stop at a FC switch.

I know I'm probably not answering your question directly...I just want to clear any misconception that all DAS is slower than SAN. It all depends on how it's done. Regarding your numbers...71K KBps is nearly 600Mbps. You couldn't expect too much more connecting to a 1Gb iSCSI SAN. That's a very simplistic comparison, however.

BTW...what's the purpose of the cross over cable? Shared storage is pretty much the most important requirement for the enterprise features of ESX...without it there is no HA, VMotion or DRS.

Hi Greg,

Without taking cache into account, these will be around about the "worst" number you will see with random write/read patterns.

15K RPM SAS disk = ~147 IO/s
10K RPM SAS disk = ~112 IO/s

So, multiply your disk qty by your I/O to get your IO total.... for you 8 x (for 10k) 112 = 896 IO/s
take your I/O size for your application.. (E.G. 8k IO size).. multiply by your IO/s total E.G. 8k x 896 = ~7.1MB/s.

So, using 10k disks in your situation, the "worst" performance you should see with a totally random I/O pattern is ~7.1MB/s.
With more sequential I/O where your cache can be utilized efficiently you will see dramatic performance increases.

If you want to get some numbers for comparison, check the following performance thread and run your own test using the included IOmeter test pattern.
http://communities.vmware.com/thread/73745

Standard things to check for disk performance is your controller "write cache" (write back) setting, make sure they are enabled (if you have a battery backup).

hope this help

Bernie
http://sanmelody.blogspot.com

The local storage I am referring to would be the 8 drive bays on the SUN server. I am concerned that 1TB spanned RAID 5 over 8 spindles could result in Disk I/O issues with 15 virtual machines accessing this storage simultaneously. I am not sure the speed of the SAS drives.

I have only used FC storage in environments I've worked in, with the exception of a MSCS setup I have. I also do not have their current disk space used, but obviously doing 4 RAID 1+0 mirrors and presenting 4 VMFS volumes of 146GB would offer a much higher level of performance, however present limitations and overhead in order to work within the constraints of 4x146 vs. even a setup that was 5x146 RAID; both offering ~584.

This setup was architected somewhat on the fly and has obviously broken some best practices, however at the presnet time, SAN storage is not an option for this project. The crossover cable is so the nodes on each host server can communicate with one another without any hops or network contention, similar to a heartbeat cable in a MSCS; granted all guest servers are running a SUSE Linux OS.

Thanks for the thoughts and input. The more research I do, I get the feeling that this storage setup could be the root of some problems, while I have identified other architectural question marks in the environment.

Thanks
~grc

Bernie

Great statistics and information. This is along the lines of what I was looking for in order to properly assess this environment. I appreciate the feedback.

~grc

Hi Greg

Its a hard decision deciding few V's many.........
Most of the time I find that primary reasons for many...are as follows...

1. Seperate VM volumes... E.G. OS, Data, DB's...and so on..
2. More volumes to split across SAN controllers / ports.
3. More volumes to reduce SCSI reservation issue's....

If you have time...couple of days....spend it to test different configurations...but if it was me...

1. I would leave the 8 drives as one RAID pack, not worth wasting drives capacity for multi RAID packs..and you would be talking about small 4 drive or less configs...
2. I would definitely create multiple Logical disks on the one RAID pack.as a minimum I would create 2 but not sure if I would bother with anymore than that...

You haven't got much capacity and disks to play with, so best not to over complicate things...There will be no point in creating to many logical disks, in the end they are on the same underlying disks...
E.G..... Moving data from one LD to another on the same physical disks....don't expect anything great from a performance point of view....

Anyway...that's my take on things..

hope that helps

Bernie.
http://sanmelody.blogspot.com

This document was generated from the following thread: ESX All Local Storage environment - performance degradation?

Servie console vmotion network failover and diagram

Thu, 19 Jun 2008 17:40:31 GMT

I have been charged with developing our VMware enterprise architecture. To that end I have produced the attached document.

I have only done the VMware fast track training so my proposals come a purely academic angle

Does anyone have an opinion of the decisions I have made?

Regards

Mike

Your decisions are good, for the most part. I did see that you had management and vmotion on individual NICs. I would recommend having redundant paths to both of these segments as well as your production traffic. If you're not going to be running huge amounts of I/O, which I'm anticipating you are not, then dedicting 4 ports to your prod network is a bit overkill. Also, if you're only using ports off the 1 4-port card, then you're setting yourself up for failure in case that card fails.

These servers will include 2 on board, and 4 on your card, giving you 6 total.

I would use 2 for vmotion, 2 for mgmt, and 2 for prod network. This should give you optimal redundancy, and should still give you sufficient network bandwidth.

Your physical network environment will come into play as well. If you're using multiple pSwitch's, then make sure your teamed NICs go to a separate pSwitch each.

Good luck, and welcome to the forums.

-KjB

At first glance that all looks pretty well thought out to me. 1 question, why are you deciding to use iSCSI for your SAN instead of fibre?

mike, I will critic your proposal and give you answers via Private Message

Tom Howarth
VMware Communities User Moderator

Tom any chance you could post them here for us all to have a read of?

Hello,

Welcome to the forums..... A few concerns...

You have redundancy for SC, iSCSI, and triple redundancy for your Production network but not for your vMotion network. If it was me, I would move one of the Production pNICs for purely redundant vMotion network. Also, you really should have 2 pSwitches for Management and vMotion for best redundancy of network capabilities. Unless these are all VLANs on multiple switches.

Not sure if the triple redundancy is to provide any thing special.

The other item is the use of ESXi... It has some security issues, and you may wish to read the Forum for posts on why to not use ESXi. If you are not all that concerned with security its a fine solution. Or you find the restrictions it imposes acceptable... Your plan should at least address all these concerns, and security seems to not be a part of the plan. It should be from the beginning.

Best regards,
Edward L. Haletky
VMware Communities User Moderator
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education. CIO Virtualization Blog: http://www.cio.com/blog/index/topic/168354, As well as the Virtualization Wiki at http://www.astroarch.com/wiki/index.php/Virtualization

The R805 is one of Dells first Virtulisation optimised servers it has 4 onboard GigE ports.

The only areas where I have used multiple pSwitches is on the production and storage networks. I was happy to have downtime on the Management network in the event of a switch failure as I can swap that if I need to with no loss of service on the production network, but I guess for the price of a switch I might double up.

Do I really need (as the VMware training material suggests) seperate networks for vMotion and Management?

WGardiner,

As we need to put in two systems it's primarilly a cost issue, the expense in a redundant switch fabric and hba's will push the budgets somewhat.

Also we have no experience in Fibre Channel

The single NIC for vMotion was a recomendation by the trainer at the VMWare training as he said that the only risk was a loss of vMotion but I see a scenario now when the NIC with the vMotion port fails and you need to put it into maintanance to repair but can't vMotion off as the port is down.

The triple redundancy was mainly because I wasn't sure what else to do with it, I was under the impression that rather than redundancy the outbound trafic would be load balanced over the 3 NICs

I was led to belive that ESXi is actually more secure? Whist at VMware in Frimley they said that if we had no existing infrastructure then ESXi was the thing to go for as it was more secure (no local service console) and was "The future"

Maybe it's a bit too soon to go for ESXi?

Unless you change the default behavior, loss of the management network, means no service console, means isolation response, which will shut down your vm's in an attempt of HA to kick in and failover your vm's to another host. You can modify your isolation response to leave the vm's up and running, but having a redundant service console would make just as much sense.

You don't typically have to have separate networks for management and vmotion, and to that end, you can use a teamed set of pNICs for both vMotion and management. Just use one as active/standby for management port, and then use standby/active for the vMotion port.

That way, you will be fully redundant.

-KjB

I see a few things that need to be taken into account.

1) You appear to be using iSCSI for your ESX storage. ESX requires iSCSI (and all NAS traffic) to use the VMKernel nics. From the diagram, it looks like you will have three VMkernel nics, with one being used for Vmotion. Now, you can use multiple VMkernel nics, but make sure you only use one active on your storage pair. ESX has had trouble handling multiple network paths on VMKernel networks.

2) In your two node cluster - you have what appears to be a 'cross-over' between your two ESX hosts for VMkernel. You need to specify a gateway device that both members can ping. This is important for determining if your ESX servers go into '*isolation mode*'. If you loose ability to hit your gateway, both members of your cluster can shutdown, and you loose any HA built in. (This gateway to ping, and isolation mode will affect all servers. I just note it here, because of the cross-over.)

3) I think you are hitting over-kill on the number of nics used, unless you are not using vlan tagging on the vswitches. Multiple vlans can be run on a single active/active trunk, and I haven't seen where one ESX host can fill a 2Gb ethernet link (2x1Gb).

-Andrew Stueve

Thanks Andrew,

2) I thought the "pinging of the gateway" occured over the management lan? I had planned to set the gateway for HA to be the virtual center server

3) We don't use vlan tagging, We've only ended up with 2xGigE for redundancy, I doubt we will be likeley to hit the limits.

We are probably going to end up with iSCSI HBA's which might free even more NICs I think we'll stick with the onboard 4 and the aditional 4 for scalability/redundancy

First off, nice writeup!

I don't see a connection between the 2 iSCSI switches. You will need the switches connected with multiple Gb links (LACP/PAGP), there is nothing stopping EthX on the ESX server from wanting to go to eth2 on the EQL boxes after it's been redirected away from the group IP addr. If that needs to traverse from switch 1 -> 2 the iSCSI connection will not work.

One other thing, since you will have a SAS and SATA EQL group you will want to put them in separate pools in the group. The current EQL firmwares recommend against mixed speed drives in the same pool.

Ben

Thanks for your feedback ben,

You make a good point with the switches I had forgotten that.

It's an interesting point about the EQL boxes needing to be in differnt pools as the Tech support from dell didn't seem to know that.

They also didn't know how long it would be before the firmware's where upgraded so that both Storage controls are active (at the same time). Any ideas?

Hi Mike,

There isn't a hard stop when attempting to add mixed speed storage into a singular pool. You will get a warning message and when the data is spread across 2 members you'll have 50% of your volume on 7.2K and 50% on 10K. That's asking for trouble if you run into any performance issues.

As for the dual-active controller question, I think that you should plan on utilizing the current active/passive features for the foreseeable future.

Ben

Hello,

If you have a spare pNIC then adding it to your vMotion network and adding some redundancy to the management switch would be my main changes to the design. Redundancy should be paramount as Kjb007 has stated, lack of access to the SC ports means no management capability, which includes vMotion.

Best regards,
Edward L. Haletky
VMware Communities User Moderator
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education. CIO Virtualization Blog: http://www.cio.com/blog/index/topic/168354, As well as the Virtualization Wiki at http://www.astroarch.com/wiki/index.php/Virtualization

first thing I noticed was a confusion as to what version of ESX you are going to deploy, you mention ESXi and foundation edition in the same paragraph. I becomes apparent as we read on but use the correct nomaclature.

see first paragraph page 2 ESXi is the version that can be installed on a Flash device, however it only has experimental HA support and no other higher function such as DRS and Vmotion again clairifaction is requried as you go on to mention VMotion networks, VCB and Service Consoles.

If you are having a phyiscal server for VC then consider a lower spec machine. also unless there is complete phyiscal seperation between the two networks make sure that one of your VC's always deploys Guests so as to prevent to posibility of a duplicate MAC address being generated.

are you sure of the ability of load balance the two SAN's as they have disparate Storage capacity, (I am not familiar with this particular manufacture)

Initial Split deployment design

Consider two cables to your VC in your production environment.

I would remove one of the NICs to your VMnetwork and add it to your VMkernel network to gain resilance.

The rest of the issues have been covered by the other posters

Tom Howarth
VMware Communities User Moderator

ESXi supports DRS and VMotion, and since ESX 3.5.0 Update 1 it also supports HA.

I stand corrected on DRS and Vmotion, but what about VCB

Tom Howarth
VMware Communities User Moderator

To my knowledge ( and I have yet to see anything to the contrary ) VCB is supported in ESXi / ESX embedded. As far as I know all versions of ESX 3.5.0 / installable / embedded should have feature parity. I know that HA was an odd item that was left out of the support matrix for ESXi when 3.5.0 came out, and that there are no CIM providers currently in ESX 3.5.0 but I believe the rest of the features/support are the same.

I didn't think I had mentioned foundation edition anywhere in the document?

"VMware ESXi 3.5 Enterprise has been selected and will be installed onto a SD card or USB stick inside the
servers."

This has been selected due baised on VMware recomendations, They see it as the future direction for there product line. As the previous poster has stated, if you buy the enterprise licence you can deploy either ESX, ESXi Installable or ESXi embedded. The trainer on the fast track VMware training courses claimed that VCB was supported.

That said other coments and further research I have conducted suggests that it might still be a little earlly to adopt ESXi.

Dell claimed trher where no problems load balancing between EqualLogic SANs of differnt sizes or Spindle speeds however one of the other poster has reported otherwise, I will further evaluate this decision once we are in a position to procced with the combined deployment.

I think I will now use the same virtually teamed and physically redundant GigE infrastrcuture for both vMotion and Management, does anyone percieve any problems with this?

you are correct you did not, however you did mention ESXi and Enterprise in the same sentence. the two are from diferent product sets

Also ESXi 's management is not a easy. the RCLI is not a feature rich. it does not even include a kill command to give the ability to shut down a hung VM from the command line

Tom Howarth
VMware Communities User Moderator

If you use the hackable features of "unsupported" command while pressing ALT+F1 and with administrator password and uncomment out # ssh part of the line from /etc/inetd.conf file you will be able to use the kill command and fully adminstrative service console tools.

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

Regards,

Stefan Nguyen
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant

Small, but I wanted to contribute...

Remember to open the ports on the ESX firewall for iscsi.

Run VC on physical, though vm is supported, based on my experiences I wouldn't do it.

Kaizen!

This document was generated from the following thread: Peer Review of ESX Architecture

Teaming NICs in ESX - configuration

lkubin@coma.cz — Wed, 09 Apr 2008 23:35:41 GMT

This article is a step-by-step guide to setup a NIC teaming both on ESX server and network switch to increase network throughput for applications.

Read the full article at my Kukacz Notes blog.

Vmotion how to?

Sun, 06 Apr 2008 12:10:14 GMT

steps

capacity bottlenecks

alexbakman@yahoo.com — Thu, 27 Mar 2008 16:16:45 GMT

Hey guys

If you are interested in beta testing a Virtual Appliance for finding capacity bottlenecks in your environment check this out

http://www.vkernel.com/products/CapacityAnalyzer/

It also monitors capacity trends and alerts via email and snmp

Full diclosure: This is my brainchild. I would love to hear your feedback

Alex

Profiler for VMware: Visibility for Physical and Virtual Worlds

Mon, 24 Mar 2008 21:31:13 GMT

Monitoring Products for VMWare ESX 3.5

Tue, 04 Mar 2008 14:51:32 GMT

I am current investigating possible monitoring tools for ESX 3.5

I know about the functionality of VMWare Virtual Centre, Vizioncore vCharter and Veeam Reporter

Are there any other reporting tools out there that people use and think are good.

Thanks

John

Yea I'm currently researching & evaluated a few products. I'm looking now at from companies Veeam, Hyperic, FastScale. My recommendations are on hold til further testing. I'm definately would be interested in what others are using in production.

Also know about Veeam Monitor aswell.-----
http://www.eginnovations.com/web/vmware.htm

Monitoring Solutions for VMware Server Farms

RParker,

How do you like eginnovations? I see they do more than just vmware. It would be nice to have a centralized monitoring of all systems and application on top of those systems. Let me know what you think so maybe I can include it in my evaluation. We are currently using What's Up Gold (installed prior to my arrival). I'm looking for a better solution overall.

Thanks

Luis

Hello,

What you want to monitor any why you want to monitor may direct you towards one product over the other. For example:

Monitoring hardware could lead you to one of the vendor packages (HPSIM, Dell Openmanage, IBM Director, etc.)

Needing Chargeback for resource utilization could lead you to vCharter from Vizioncore

Centralized up/down server monitoring could be handled by just about anything out there.

Do you want NOC like functionality? Accounting? Auditing? Security? etc. Without knowing which, we end very quickly to just discussing general tools with no real specifics. For general monitoring Virtual Center has quite a bit as well. So what do you wish to monitor? What are you looking to get out of it?

Best regards,
Edward L. Haletky
VMware Communities User Moderator
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education. As well as the Virtualization Wiki at http://www.astroarch.com/wiki/index.php/Virtualization

We use Nworks management pack for Microsoft Opereations ... so you you use OM i can reccomend it -http://www.nworks.com/vmware/

Take a look at netuitive or virtugo:

In my opinion netuitive is far beyond the other monitoring solutions, because not only reacting , it is proactiv and selflearning.

http://www.netuitive.com/

http://www.it-director.com/technology/applications/content.php?cid=10042

http://www.byteandswitch.com/document.asp?doc_id=119121

http://virtualization.sys-con.com/read/507887.htm

http://www.uxcomm.com/

This document was generated from the following thread: Monitoring Products for VMWare ESX 3.5

What are my options for Backup of a VM

Wed, 06 Feb 2008 10:49:51 GMT

How do I backup my Virtual Infrastructure?

Some initial statements

Backup is for the sole purpose of restore. People often don't think about restoration until the end, if all. IMHO you should start with restore and move backwards from that.

Additionally the level of restore speed, quite often called the Recovery Time Objective (RTO) and how old the last backup is allowed to be, quite often called the Recovery point Object (RPO) will also define on a per machine basis how a backup architecture is structured.

Often a hybrid scheme is required. One method is not going to suit every system, because they have different restore requirements or available methods of backup.

Options for backing up Virtual Machines

As described by the Virtual Machine Backup Guide there a few methods to backup a VM. Here is a summary.

Method	Description	Benefits	Problems	Consistency	Restoration
Backup agent inside the VM.	Often an approach as people start to consolidate existing physical servers. Run the agent for you backup software inside the VM which sends the data to your backup server.	The agent can be application aware and perform functions not possible with other methods. For example application quiescing, log truncating or backup in a non-filesystem but application aware form, such as mailboxs. An easy transition from an existing backup scheme, no changes Can selectively choose what to backup or exclude	Requires a license for each VM. Consumes production network traffic (unless separate backup network implemented) Does not capture the entire machine state (in most software) A processing load is placed on the host	File system and application through agent	Performed directly back into the VM through the agent Mounted/restored onto backup server
Backup agent within the service console of ESX	3rd party backup software or scripts are installed into the service console which interact with the VM files to send them to a backup server or local device	A full VM backup can be performed	Relies on the service console Software must be maintained on the ESX server A processing load is placed on the host, can be significant Snapshots have to be executed, file system consistency only	File system through VMware tools and snaphot	Transfer the files back into an ESX host
Backup applications which integrate with VC and ESX	Vendors provide applications which interact with the service console to extract data without requiring software installed Examples include vRangerPro and esXpress	A full VM backup can be performed Being VMware aware backup polocies can be applied on VC elements such as folders of VMs Often integrate with VCB Often provide support for incremental backups	Requirs acess to service console so no support for 3i by vendors yet. Consumes production network traffic (inless separate backup network implemented)	File system through VMware tools and snapshot. Some have application through agent in VM	Transfer the files back into an ESX host mounted/restored onto backup server
Backup at the storage layer	Using storage vendor techniques to present the data for backup, such as SAN snaphots	Good for RDM Good for very large amounts of data	Not granular, backup of the whole LUN Difficult to co-ordinate the queisce of all VM disks on the LUN	Crash consistant unless snapshots co-ordinated. File system with VMware snapshots	Mounted/restored onto backup server
VMware Consolidated Backup	VCB can perform full image backups of VMs via the storage system or over the network. Either a full VM dump or presentation as read only mount point for backup by other backup software (Windows VM only)	Works for all OS types No cost for additional agents Comes included with all version except 3i Integration modules for most leading backup vendors The only way to backup full VMs in 3i	No compression of the backup images No incremental on full backups No scheduling or audit logs Can require a large amount of space for the holding tank	File system conistant with VMware tools. Pre/post scripts support for applicicaton integration (but roll your own)	VMware Converter

Running your backup server as a VM

The good bad and ugly of running a backup server

Backup vendors

List of backup vendors and links

vRanger Pro http://www.vizioncore.com/vRangerPro.html
esXpress http://www.esxpress.com/
Symantec Netbackup v6.5.1 http://entsupport.symantec.com/docs/293350
HP Data Protector V6.0 Integration
Veeam Backup http://www.veeam.com
Tivoli Storage manager http://www.ibm.com/tivoli

Specific application suggestions

Suggestions on specific applications

MS Exchange

Exchange ideas

MS SQL

SQL ideas

Other considerations

Don't forget to consider any requirements for backup of any physical system outside your VMware infrastructure.
Problems reported with the quiesce software used in tools that can cause some corruptions. (to be expanded)

References

Virtual Machine Backup Guide http://www.vmware.com/pdf/vi3_35/esx_3/r35/vi3_35_25_vm_backup.pdf

Which RAID type should I use?

Tue, 29 Jan 2008 10:29:59 GMT

This document is not a summary of storage best practices. Its purpose is to assist in the answering of the question, "What RAID type do I configure my VMFS LUN as?" which is often asked in regards to local storage.

Summary of the different RAID types

Type	RAID 0	RAID 1	RAID 1+0	RAID 5
Method	Striping	Mirroring	Mirror and striping	Stripe with Parity
Description	The data is striped across all the disks in the set. Not true RAID because it is not redundant.	Exact copies of the data are kept across paired disks or disk sets.	A striped array whose segments are a mirror set. Not to be confused with 0+1	The data is striped across all the disks in the RAID set, along with the parity information needed to reconstruct the data in case of disk failure.
Performance	Good. All IO is spread and no parity overhead	Twice the read rate but writes occur twice	Combines speed efficiency of the RAID 0 (stripe set) with a fault tolerance of RAID 1 (mirror).	High read and medium write performance. A good balance of performance and availability.
Typical use	Data you can afford to loose, such a pre-production, image/video editing.	Data requiring high availability, such as accounting or payroll.	Transactional type data such as database logs.	File and application servers. Typically everything that does not have a specific requirement.
Parity Calculation	None	None	None	Yes
Redundancy	None	Yes	Yes	Yes
Minimum Disks	2	2	4	3
Storage Overhead	0%	100%	100%	Capacity of one disk in the set. 33% for a 3 disk set.

Other elements to consider are

Hot spare disks can often be configured to provide a space to rebuild a redundant disk without attendance.
Some types of SAN behave differently under the hood and you should reference the vendor documentation and best practices. For example the HP EVA works quite differently and some vendors have RAID 6.
For optimal performance make sure VMFS is the only partition on LUN and make sure the LUN maps to it's own set of physical disks if possible.

Document maintenance

This document is open for public maintenance in order to be a living document maintained by the community. Please update as required with the following guidelines.

Attempt to keep formatting consistent and simple.
Mark revisions as required. Minor edits don't need to be new versions.
Post a comment describing changes to the document rather than inserting revision information into the document in order to keep the document uncluttered.
Try to stick to the facts rather than opinions.
References to further reading are helpful.

List of VMware Whitepapers

Thu, 24 Jan 2008 13:32:09 GMT

VMware provide a list of Technical Whitepapers at http://www.vmware.com/solutions/whitepapers.html

The following document details each of the documents currently listed with a link to the PDF and the detailed description, all in one listing. This way when you search on VMTN you also have a chance of finding the white papers which are relevant to your topic.

Compiled by Rodney Haywood(Rodos), VMware Practice Manager, Oriel Technologies, Sydney Australia, 24Jan2008

General

Virtualization Overview

Get an overview of the different types of Virtualization that exist in the industry today.

VMware Paravirtualization

Understanding full Virtualization, Paravirtualization, and Hardware Assist.

Virtualization: Architectural Considerations and Other EvaluationCriteria

Examine the architectural issues, solution support and enterprise readiness required when deploying a virtualization solution.

Building the Virtualized Enterprise With VMware Infrastructure

Learn how to create a scalable and responsive IT organization with VMware Infrastructure.

Addendum to Microsoft Virtualization Licensing and Distribution Terms

Read about new developments and additional details related to the white paper on Microsoft virtualization licensing and business practices.

Microsoft Virtualization Licensing and Distribution Terms

Learn about Microsoft's licensing and business practices for virtualization and how Microsoft restricts usage of virtualization.

Server Consolidation

Building the Virtualized Enterprise With VMware Infrastructure

This paper clarifies how adopting VMware Infrastructure - the combination of server, storage and networking virtualization technologies -as a fundamental IT strategy helps organizations achieve sustainable competitive advantage.

Reducing Server Total Cost of Ownership (TCO) With VMware

This paper explains how leading VMware customers have achieved considerable improvements in IT infrastructure TCO by virtualizing their datacenters. It also indicates how prospective VMware customers can estimate their own potential TCO and return on investment (ROI).

Using ESX Server and VirtualCenter to Reduce Oracle Real Application Clusters Deployment Costs and Cycle Times

This paper describes the steps necessary to clone Real Application Cluster nodes with ESX Server and VirtualCenter.

Improving Economics of Blades using VMware

This marketing whitepaper emphasizes cost advantages and operational benefits of deploying VMware software on blades platform.

Best Practices on ESX Server

This technical whitepaper covers best practices of deploying virtual infrastructure.

20:1 Consolidation and Server Virtualization Solution

These whitepapers developed in collaboration by Intel, IBM, and VMware describe technical architecture and business advantage of server consolidation with VMware virtual infrastructure on enterprise class Intel architecture servers.

Server Consolidation with VMware and HP ProLiant

This whitepaper developed by HP describes the advantages of deploying VMware software on HP servers.

IT Consolidation with Linux

This high-level whitepaper developed by HP describes various methods of consolidating Linux environments from virtualization to superdome.

Using VERITAS NetBackup DataCenter 4.5 with ESX Server

This whitepaper covers best practices of protecting data deployed on virtual infrastructure with VERITAS NetBackup software.

OS Migration

OS Migration and Legacy Application Support Using VMware Virtual Machine Software.

Recentralizing Server Sprawl Through VMware

This whitepaper covers best strategies to cost savings.

VMware: Tool for Server Consolidation

Summary from D.H. Brown on Server Consolidation.

Development and Testing

Develop, Test, and Deploy Faster with VMware Software

This high level whitepaper describes several scenarios where virtual infrastructure dramatically improves test and development optimization.

Streamlining Patch Testing and Deployment with LANDesk

This whitepaper describes a solution for patch management based on integration of LANDesk Management Suite and VMware GSX Server.

Business Continuity

Transforming Disaster Recovery

This paper covers how to use VMware Infrastructure for rapid, reliable and cost-effective disaster recovery.

Building the Virtualized Enterprise With VMware Infrastructure

This paper clarifies how adopting VMware Infrastructure - the combination of server, storage and networking virtualization technologies -as a fundamental IT strategy helps organizations achieve sustainable competitive advantage.

Making Your Business Disaster Ready with Virtual Infrastructure

This paper describes several examples of improving the costs and disaster response time with VMware virtual infrastructure.

VMware ESX Server Backup

This paper covers how ESX Server provides many possibilities for backup, restore and disaster recovery in data center management.

Desktop Manageability and Security

Manage and Standardize Guest PC Access with VMware ACE

This solution datasheet describes the challenges of guest access and solution using VMware ACE.

Manageable and Safe Remote Access with VMware ACE

This solution datasheet describes the challenges of remote access and solution using VMware ACE.

VMware Virtual Desktop Infrastructure (VDI) Server Sizing and Scaling

This white paper describes the testing methodology, results, and analysis and sizing guidelines for setting up Virtual Desktop Infrastructure using VMware Infrastructure 3.

Addressing Desktop Challenges with a VMware Virtual Desktop Infrastructure (VDI)

This solution datasheet describes the benefits of using VMware Virtaul Desktop Infrastructure to address challenges in desktop management for TCO, security and resource utilization.

Using VMware Virtual Desktop Infrastructure (VDI) for Offsite Desktop Management

This solution datasheet describes the benefits of using VMware Virtaul Desktop Infrastructure to address challenges in offsite desktop management.

Virtual Desktop Infrastructure (VDI)

This white paper discusses how:

VDI simplifies desktop administrative and management tasks
Users access virtual desktops running in the data center
Technology supports both PCs and thin clients
Desktop security and data protection are centralized
Access can be easily extended to remote users

Software Lifecycle Automation

VMware Software Lifecycle Automation Solutions

This paper covers VMware software lifecycle automation solutions, which support a wide range of deployment scenarios from individual users to large multi-site teams.

Virtual Lab Automation

Across the board, enterprise software development organizations are grappling with server sprawl, setup and provisioning overhead, and costly system failures. This paper discussed how Virtual Lab Automation (VLA) can substantially mitigate these problems, and thus become a new and critical “best practice” component in any software development and test environment.

VMware Technical Resource documents listing

Wed, 23 Jan 2008 04:38:01 GMT

VMware provide a list of Technical Resource Documents at http://www.vmware.com/resources/techresources/cat/91

The listing provides little search capabilities and it can be a challenge to find a document relevant to a topic of research. The following document details each of the documents currently listed with a link to the PDF and the detailed description, all in one listing. Documents are listed in order of revision.

Compiled by Rodney Haywood(Rodos), Enterprise Architect - Virtualisation, Alphawest Services, Sydney Australia

206 documents listed.

Dell EqualLogic VMware Virtual Desktop Infrastructure
by Dell on 01/25/2009 @ http://www.vmware.com/files/pdf/partners/dell/dell-vmware-solution-brief-r2final08q4.pdf

This solution brief explains how organizations can use VMware virtualization technologies to provide a virtual desktop infrastructure.

Dell EqualLogic VMware® View 3
by Dell on 01/25/2009 @ http://www.vmware.com/files/pdf/partners/dell/dell-vmware-desktop-wp-r208q4.pdf

This paper assumes that the reader already has a good understanding of desktop virtualization

and has reached the point where they are ready to begin sizing their server and storage architecture.

Address Desktop Challenges with VMware View and NetApp
by NetApp on 01/24/2009 @ http://www.vmware.com/files/pdf/partners/netapp-tap-desktop-solution-brief.pdf

NetApp helps you realize the full potential of your VMware environment by addressing the common storage challenges associated with virtual desktop solutions.

Comprehensive Virtual Desktop Deployment with VMware and NetApp
by NetApp on 01/24/2009 @ http://www.vmware.com/files/pdf/partners/netapp-vmware-view-wp.pdf

VMware offers an end-to-end solution called VMware® View, the next generation of VMware VDI, that allows organizations to provide corporate end users with access to virtual desktop machines that are hosted in a central data center.

xpnet Performance White Paper
by xpnet on 01/21/2009 @ http://www.vmware.com/files/pdf/xpnet_Performance_Review_of_AppVirt_Solutions.pdf

Application Virtualization 2008-2009:

Assessing the Architectural and Performance Characteristics of Four Leading Windows Application Virtualization Solutions

Java in Virtual Machines on VMware ESX: Best Practices
by VMware on 01/21/2009 @ http://www.vmware.com/files/pdf/Java_in_Virtual_Machines_on_ESX-FINAL-Jan-15-2009.pdf

This paper discusses best practices for running Java-based software in VMware ESX virtual machines. These guidelines will help you to get the best from your Java applications and application servers when you run them on VMware Infrastructure 3.

VMware ThinApp 4 Reviewer’s Guide
by VMware on 01/19/2009 @ http://www.vmware.com/files/pdf/thinapp_4_reviewers_guide.pdf

VMware ThinApp 4 Reviewer’s Guide

Performance of Virtual Desktops in a VMware Infrastructure 3 Environment
by VMware on 01/20/2009 @ http://www.vmware.com/files/pdf/vdesk_scaling.pdf

This paper examines the performance of virtual desktops running a typical mix of interactive applications on VMware ESX 3.5 Update 2. These include office application tasks such as editing documents, spreadsheets, and presentations, as well as browsing the Internet and reading documents. Results show the effect on performance as workload is scaled up from 16 to 160 virtual machines.

VMware View Reference Architecture Kit
by VMware on 12/09/2008 @ http://www.vmware.com/files/pdf/resources/vmware-view-reference-architecture.pdf / http://www.vmware.com/files/pdf/resources/vmware-view-profile-virtualization.pdf / http://www.vmware.com/files/pdf/resources/vmware-view-xp-deployment-guide.pdf / http://www.vmware.com/files/pdf/resources/vmware-view-ns20-deployment-guide.pdf

This reference architecture kit is comprised of four distinct papers written by VMware and our supporting partners to serve as a guide to assist in the early phases of planning, design and deployment of VMware View based solutions. The building block approach uses common components to minimize support costs and deployment risks during the planning of VMware View based deployments.

Included in this kit are the following materials:

VMware View Reference Architecture
Guide to Profile Virtualization
Windows XP Deployment Guide
Storage Deployment Guide for VMware View

Storage Design Options for VMware Virtual Desktop Infrastructure
by VMware on 11/29/2008 @ http://www.vmware.com/files/pdf/resources/WP_Storage_Design_Options_For_VMware_VDI.pdf

This paper provides information on technical concepts related to storage implementations in a VMware ® Virtual Desktop Infrastructure (VDI) environment.

Using IP Multicast with VMware ESX 3.5
by VMware on 12/01/2008 @ http://www.vmware.com/files/pdf/technology/esx35_ip_multicast.pdf

This paper explains the operation of IP Multicast in ESX 3.5 and the considerations and best practices for deployment.

VMware Distributed Power Management: Concepts and Usage
by VMware on 11/24/2008 @ http://www.vmware.com/files/pdf/DPM.pdf

VMware Distributed Power Management (DPM) saves power in a VMware Infrastructure 3 cluster of ESX hosts by consolidating virtual machines onto fewer hosts and powering hosts off during periods of low resource utilization, and powering hosts back on for virtual machine use when workload demands increase. DPM is an optional add-on to VMware Distributed Resource Scheduling (DRS).

This paper provides a technical overview of DPM operation in the VirtualCenter 2.5 / ESX 3.5 release. It is intended for VMware partners, resellers, and customers who want detailed information on DPM functionality in that release.

SQL Server Workload Consolidation
by VMware on 11/25/2008 @ http://www.vmware.com/pdf/SQL_Server_consolidation.pdf

Database workloads are very diverse. While most database servers are lightly loaded, larger database workloads can be resource-intensive, exhibiting high I/O rates or consuming large amounts of memory. With improvements in virtualization technology and hardware, even servers running large database workloads run well in virtual machines. Servers running Microsoft's SQL Server, among the top database server platforms in the industry today, are no exception.

Performance of Rapid Virtualization Indexing (RVI)
by VMware on 11/13/2008 @ http://www.vmware.com/pdf/RVI_performance.pdf

AMD recently introduced its second generation of hardware support for virtualization, incorporating MMU virtualization called Rapid Virtualization Indexing (RVI). Hardware support for MMU virtualization can improve performance, particularly for MMU-intensive workloads.

VMware ESX 3.5 leverages this RVI support in AMD processors. This paper compares the performance with and without RVI of a number of industry-standard benchmarks and microbenchmarks running in VMware ESX 3.5, Update 2 on AMD Opteron 8384 ("Shanghai") processors.

10Gbps Networking Performance on ESX 3.5 Update 1
by VMware on 11/04/2008 @ http://www.vmware.com/pdf/10GigE_performance.pdf

10 Gigabit Ethernet is expected to replace Gigabit Ethernet and become the dominant Ethernet standard in the next few years. VMware ESX now supports a number of 10 Gbps network cards and allows multiple virtual machines to share a single physical NIC. This paper presents results from our single-virtual machine and multi-virtual machine network throughput experiments that show ESX can easily reach line rates on 10 Gbps links. The paper discusses how Jumbo Frames influence networking performance, both on the receive and the transmit paths. The paper also presents the results of our scalability experiments in which up to 16 virtual machines share a single 10 Gbps physical NIC and discusses the allocation of bandwidth to different virtual machines.

VMware Update Manager 1.0 Performance and Best Practices
by VMware on 11/03/2008 @ http://www.vmware.com/pdf/vum_1.0_performance.pdf

This paper provides measurements of VMware Update Manager (VUM) 1.0 performance and gives recommendations about VUM deployment and configuration.

Installing and Configuring Linux Guest Operating Systems
by VMware on 10/30/2008 @ http://www.vmware.com/files/pdf/linux_install_config.pdf

This technical note describes installing, configuring, updating, and administering Linux guest operating systems in virtual machines running on VMware Infrastructure 3 version 3.5. In addition, this note includes a collection of useful tips and tricks in fine-tuning your Linux virtual machines. Although the recommendations in this paper apply to most Linux distributions, they are tailored specifically to Red Hat Enterprise Linux 5. Linux administrators can use this paper as a source for guidelines when building and maintaining Linux virtual machines in their VMware Infrastructure environments. Some working knowledge of VirtualCenter 2.5 Update 2, ESX 3.5 Update 2, and Linux operating systems is required.

Best Practices for Patching VMware® ESX/ESXi
by VMware on 10/09/2008 @ http://www.vmware.com/files/pdf/esx_patching_best_practices.pdf

This best practices document gives a brief explanation of patching and the different mechanisms for applying patches for the VMware ESX 3.5 product line. This paper does not focus on the differences in detail between proactive patch management and reactive patch management. Instead, it focuses on the ESX patching model, patch packaging, and deliverables – key topics you need to understand to maintain your ESX systems.

VMware Consolidated Backup ‐ Partner Integration Guide
by VMware on 09/29/2008 @ http://www.vmware.com/files/pdf/vcb_partner_integration_guide.pdf

This guide offers information on VMware Consolidated Backup integrations developed by VMware Technology Alliance Partners. The functionality described here is supported by the partner Support organization directly rather than by VMware Global Support Services.

VMware VDI Storage Considerations
by VMware on 09/29/2008 @ http://www.vmware.com/files/pdf/vdi-storage-considerations-v4.pdf

This information guide focuses on design choices for the storage environment that forms the foundation of a production VMware VDI implementation.

Implementing Microsoft Network Load Balancing (NLB) in a Virtualized Environment
by VMware on 09/25/2008 @ http://www.vmware.com/files/pdf/implmenting_ms_network_load_balancing.pdf

Network Load Balancing (NLB) is a feature of recent Microsoft Windows Server operating systems used to improve the scalability and availability of Internet server applications. This paper provides instructions for supporting NLB in unicast or multicast mode on VMware ESX.

Virtual Appliances: A New Paradigm for Software Delivery
by VMware on 09/23/2008 @ http://www.vmware.com/files/pdf/vam/VMware_Virtual_Appliance_Solutions_White_Paper_08Q3.pdf

This white paper discusses the value of virtual appliances and outlines the ways in which VMware is building a rich and diverse ecosystem around the virtual appliance model based on four key elements:

A VMware Ready Virtual Appliance Program which is an ecosystem initiative that enables customers to identify virtual appliances that follow best practices and are optimized for VMware Infrastructure.
A robust authoring tool for configuring, packaging and updating production-ready virtual appliances.
A large and diverse marketplace where vendors can list their virtual appliances online for customers to discover, download, evaluate and purchase.
A reliable, robust and proven virtualization platform for deploying and managing virtual appliances.

Virtualizing SAN Connectivity with VMware Infrastructure 3 and Brocade Data Center Fabric Services
by VMware on 09/03/2008 @ http://www.vmware.com/files/pdf/VMware-BRCD_Virtualizing_SAN_Connectivity_GA-TB-084-00.pdf

This Joint best practice technical white paper addresses the benefits of using NPIV with ESX Server 3.5. It addresses best practices and deployment considerations. Providing specific examples of deployment steps and two case for which the NPIV can be leveraged with Fabric quality of service (QoS) in VMware environments.

Best Practices Guide: Emulex Virtual HBA® Solutions and VMware® ESX Server 3.5
by VMware on 09/2/2008 @ http://www.vmware.com/files/pdf/VMware_Emulex_Best_Practices_for_Virtual_HBA_V4.pdf

This Joint best practice technical white paper addresses the benefits of using NPIV with ESX Server 3.5. It addresses best practices and deployment considerations. This paper provides specific examples of deployment steps and use cases for which the joint technology can be leveraged in virtualization environments.

Using EMC SRDF Adapter for VMware® Site Recovery Manager
by VMware on 08/28/2008 @ http://www.vmware.com/files/pdf/VMware_SRM_SRDF_bestpractices.pdf

This white paper discusses the best practices for using VMware® Site Recovery Manager with EMC® SRDF® Adapter.

Performance Counters
by VMware on 08/27/2008 @ http://www.vmware.com/files/pdf/technote_PerformanceCounters.pdf

This technical note provides additional information about performance counters available in ESX.

Timekeeping in VMware Virtual Machines
by VMware on 08/15/2008 @ http://www.vmware.com/pdf/vmware_timekeeping.pdf

This paper describes how timekeeping hardware works in physical machines, how typical guest operating systems use this hardware to keep time, and how VMware products virtualize the hardware.

VMware Management and Automation Products: Deployment Approaches and Considerations
by VMware on 08/20/2008 @ http://www.vmware.com/files/pdf/MandA_deployment.pdf

VMware Management and Automation products provide datacenter automation solutions in IT service delivery and business continuity. These products can be deployed across a shared VMware Infrastructure platform. This paper presents technical guidance and considerations when deploying these products together on a shared VI platform.

VMbook - Business Continuity and Disaster Recovery
by VMware on 08/07/2008 @ http://www.vmware.com/files/pdf/practical_guide_bcdr_vmb.pdf

This VMware® VMbook focuses on business continuity and disaster recovery (BCDR) and is intended to guide the reader through the step-by-step process to set-up a multisite VMware Infrastructure that is capable of supporting BCDR services for designated virtual machines at time of test or during an actual event that necessitated the declaration of a disaster, resulting in the activation of services in a designated BCDR site.

DRS Performance and Best Practices
by VMware on 07/28/2008 @ http://www.vmware.com/files/pdf/drs_performance_best_practices_wp.pdf

VMware Infrastructure 3 provides a set of distributed infrastructure services that make the entire IT environment more serviceable, available, and efficient. Working with VMware ESX 3, VMware VirtualCenter 2, and VMware VMotion, VMware Distributed Resource Scheduler (DRS) dynamically allocates resources to enforce resource management policies while balancing resource usage across multiple ESX hosts. This performance study focuses on understanding the effectiveness and scalability of DRS algorithms. It identifies various scenarios in which you can benefit from DRS and explains how to configure your environment to take best advantage of DRS.

Performance Best Practices and Benchmarking Guidelines
by VMware on 07/10/2008 @ http://www.vmware.com/pdf/VI3.5_Performance.pdf

This book provides tips that help administrators maximize the performance of VMware Infrastructure 3 version 3.5. A separate chapter provides guidance on benchmarking VMware ESX systems.

Data Protection Options for Exchange
by VMware on 06/25/2008 @ http://www.vmware.com/files/pdf/data_protection_options_exchange.pdf

More and more customers are deploying their mission critical applications, prominently Exchange, on VMware Infrastructure platform. This paper describes various options that customers have in protecting Exchange installation against data loss. The paper also summarizes the merits of each of the options.

Scalable Storage Performance
by VMware on 06/05/2008 @ http://www.vmware.com/files/pdf/scalable_storage_performance.pdf

VMware ESX enables multiple hosts to share the same physical storage reliably through its highly optimized storage stack and VMware Virtual Machine File System (VMFS). To gain the greatest advantage from shared storage, it is important to understand the potential bottlenecks at various layers and make the necessary configuration changes to get optimal storage performance.This paper presents the results of our studies on storage scalability in a virtual environment with many ESX hosts, many LUNs, or many of both. It examines the effects of I/O queuing at various layers in a virtual infrastructure as more and more virtual machines share the same storage. It considers the effects of SCSI reservations on virtual machine I/O performance. And it looks at ways to mitigate bandwidth bottlenecks when multiple LUNs are connected to a single ESX host. It provides recommendations you can follow to avoid overcommitting storage resources.

VMware VDM 2.1 Reviewers Guide
by VMware on 06/04/2008 @ http://www.vmware.com/files/pdf/vdm_2.1_reviewers_guide.pdf

This Reviewer's Guide provides a step-by-step process to get VDM 2.1 installed and configured for evaluation purposes.

Virtual Desktop Infrastructure: Deployment Considerations
by VMware on 06/04/2008 @ http://www.vmware.com/files/pdf/vdi_deployment_considerations_wp.pdf

This white paper is the first in a series that detail deployment considerations that should be taken into account when planning a VMware VDI deployment.

Consolidating Web Applications Using VMware Infrastructure
by VMware on 05/29/2008 @ http://www.vmware.com/files/pdf/consolidating_webapps_vi3_wp.pdf

Enterprises continue to move their technologies and services onto the Web. Today, the Web servers that provide these services are distributed across multiple systems. As the number of Web applications increases, it is very common for the number of physical systems in the data center hosting these Web applications to increase along with them. Studies from IDC, among others, describe the challenges IT managers face administering the proliferation of servers used to run Web applications. Virtualization can help businesses to consolidate their Web computing needs onto fewer high performance servers. This approach can simplify management, save operating costs, and increase the efficiency of delivering Web services.

In this paper we explore the configuration and testing of VMware® Infrastructure 3 as a consolidation platform for multiple Apache Web servers. It describes the processes and methodologies we used in the consolidation study. In addition, we describe the results of our performance testing using the industry standard SPECweb2005 workload to determine the effectiveness of this consolidation approach.

Oracle Database Scalability in VMware ESX
by VMware on 05/14/2008 @ http://www.vmware.com/pdf/Oracle_Scaling_in_ESX_Server.pdf

Database applications running on individual physical servers represent a large consolidation opportunity. However enterprises considering such consolidation want guidance as to how well databases scale using virtualization.

In this paper we demonstrate that when running multiple virtual machines with Oracle database workloads on VMware ESX, the individual performance remains close to that of the Oracle database workload in a single virtual machine, while CPU utilization scales in a near-linear fashion.

This scalability is one of the factors that makes VMware ESX the perfect platform on which to consolidate demanding, mission-critical workloads such as Oracle databases.

Implementing Virtual Provisioning on EMC Symmetrix DMX with VMware Virtual Infrastructure
by EMC on 06/15/2008 @ http://www.vmware.com/files/pdf/symmetrix_virtual_provisioning_wp.pdf

This white paper provides a detailed description of the technical aspects and benefits of deploying VMware Virtual Infrastructure version 3 on EMC Symmetrix devices using Virtual Provisioning.

Thin Client Compatibility Guide for Virtual Desktop Infrastructure (VDI) and Virtual Desktop Manager
by VMware on 06/30/2008 @ http://vmware.com/pdf/VDM_ThinClient_guide.pdf

VMware Virtual Desktop Infrastructure is an integrated desktop virtualization solution that delivers enterprise-class control and manageability with a familiar user experience. VMware Virtual Desktop Manager provides simplified management and secure provisioning of virtual desktops.

VMware Virtual Desktop Manager is tested for compatibility with a variety of thin client devices. Our goal is to support a wide variety of thin clients and virtual desktop devices used to access virtual desktops through VMware Virtual Desktop Infrastructure.

DMZ Virtualization with VMware Infrastructure
by VMware on 05/08/2008 @ http://www.vmware.com/files/pdf/dmz_virtualization_vmware_infra_wp.pdf

As virtualization of network DMZs becomes more common, demand is increasing for information to help network security professionals understand and mitigate the risks associated with this practice. This paper provides detailed descriptions of three different virtualized DMZ configurations and identifies best practice approaches that enable secure deployment.

VirtualCenter Database Performance for Microsoft SQL Server 2005
by VMware on 04/14/2008 @ http://www.vmware.com/files/pdf/vc_database_performance.pdf

VMware® VirtualCenter uses a database to store metadata on the state of a VMware Infrastructure environment. Performance statistics and their associated stored procedure operations constitute the largest and the most resource‐intensive component of the VirtualCenter database. Hence the performance of your VirtualCenter database depends upon the frequency at which you collect performance statistics and the level of detail of the statistics you store. VirtualCenter 2.5 features a number of enhancements that are aimed at greatly improving the performance and scalability of the performance statistics operations in the VirtualCenter database. The purpose of this study is to present the performance results of tests we conducted to validate these performance enhancements and to provide best practices information for configuring a VirtualCenter database. The study also provides information for sizing the server you use to host the VirtualCenter database based on these performance results. Although the new features in VirtualCenter 2.5 benefit users with any of the supported databases, the examples and performance data presented in this study are specific to Microsoft SQL Server and the paper assumes that you have a working knowledge of SQL Server.

IBM® Lotus® Sametime® Reference Architecture in a VMware® Infrastructure 3 Environment
by VMware on 04/04/2008 @ http://www.vmware.com/files/pdf/ibm_lotus_sametime_ref_arch_vi3_wp.pdf

Real-time collaboration has become a vital component of on-demand business, and IBM Lotus Sametime has quickly become an important component in many IBM Lotus Notes® and IBM Lotus Domino® environments. Typical Lotus Sametime deployments that support several thousand users require multiple servers in the physical environment, which results in high capital and operating costs. Virtualization offers unprecedented opportunities in commodity server consolidation, workload management, and return on investment.

VIRTUALIZED iSCSI SANS: Flexible, Scalable Enterprise Storage for Virtual Infrastructures
by VMware on 03/12/2008 @ http://www.vmware.com/files/pdf/partners/dell-eql-iscsi-sans-wp.pdf

This white paper describes a virtualized infrastructure that applies storage and server virtualization technologies to cost-effectively achieve a flexible, high-performance, dynamic IT infrastructure that is simple to manage and scale.

Performance Comparison of Virtual Network Devices
by VMware on 03/11/2008 @ http://www.vmware.com/files/pdf/perf_comparison_virtual_network_devices_wp.pdf

The networking performance of a virtual machine is greatly influenced by the choice of virtual network devices in the virtual machine and the physical devices configured on the host machine. ESX Server 3.5 supports multiple virtual network devices (vlance, e1000, vmxnet), each with its own usability advantages and performance benefits. It is clear that vlance is not the best choice for high-performance workloads. However, many users are still uncertain about the performance differences between e1000 and vmxnet virtual network devices.

What's New in VMware Infrastructure 3: Performance Enhancements
by VMware on 02/28/2008 @ http://www.vmware.com/files/pdf/vi3_performance_enhancements_wp.pdf

The new features in VMware® Infrastructure 3 makes it even easier for organizations to virtualize their most demanding and intense workloads. The new version of VMware Infrastructure 3 provides significant performance enhancements, including the release of VMware ESX Server 3.5 and a new ultra-thin hypervisor called VMware ESX Server 3i that can significantly

Networking Performance in VMware ESX Server 3.5
by VMware on 02/25/2008 @ http://www.vmware.com/files/pdf/ESX_networking_performance.pdf

There is a general belief that due to the extra layer introduced by virtualization code, networking performance in a virtualized environment cannot match the performance in a native environment.

Performance Characterization of VMFS and RDM Using a SAN
by VMware on 02/14/2008 @ http://www.vmware.com/files/pdf/performance_char_vmfs_rdm.pdf

VMware ESX Server offers two choices for managing disk access in a virtual machine—VMware Virtual Machine File System (VMFS) and raw device mapping (RDM). It is very important to understand the I/O characteristics of these disk access management systems in order to choose the right access type for a particular application. Choosing the right disk access management method can be a key factor in achieving high system performance for enterprise‐class applications.

Large Page Performance
by VMware on 02/14/2008 @ http://www.vmware.com/files/pdf/large_pg_performance.pdf

VMware ESX Server supports the use of large pages inside virtual machines. The large‐page support enables server applications to establish large‐page memory regions. Memory address translations use translation lookaside buffers (TLB) inside the CPU. The use of large pages can potentially increase TLB access efficiency and thus improve program performance. This study examines the performance of this feature.

Performance of VMware® VMI
by VMware on 02/13/2008 @ http://www.vmware.com/pdf/VMware_VMI_performance.pdf

VMware ESX Server 3.5 introduces support for guest operating systems that use VMware's paravirtualization standard, Virtual Machine Interface (VMI). This paper describes VMI and its performance benefits, concluding that VMI-style paravirtualization offers performance improvements for a wide variety of workloads, but that the actual performance gains depend on the nature of those workloads.

Maintaining End-to-End Service Levels for VMware® Virtual Machines Using VMware DRS And EMC Navisphere QoS
by EMC on 02/05/2008 @ http://www.vmware.com/files/pdf/EMC_VMware_NQM_wp.pdf

With the increasing deployment of VMware® ESX Server for mission-critical applications, it becomes even more important to maintain the performance of these applications in virtual machines. The complexity of virtualization, especially when server and storage resources are shared across multiple components, makes it more difficult for an administrator to detect and solve performance problems.

Using CLARiiON®’s Navisphere® Quality of Service (NQM) product and VMware’s Distributed Resource Scheduler allows you to maintain application service levels in virtual machines. This greatly minimizes the time and resources needed to manage and fix performance issues.

H2756 - Using EMC Celerra IP Storage with VMware Infrastructure 3 over iSCSI and NFS (EMC White Paper)
by EMC on 01/31/2008 @ http://www.vmware.com/files/pdf/VMware_VI3_and_EMC_Celerra_IP.pdf

With th release of VMware Infrastructure 3, virtual hardware support was extended to include the use of IP storage devices. This support enables ESX environments to take full advantage of the NFS protocol and IP block storage using iSCSI. This significant improvement profivdes a method to tie virtualized computing to virtualized storage, offering a dynamic set of capabilities within the data center and resulting in improved performance and system reliability. This white paper describes how best to utilize EMC Celerra in a VMware ESX environment.

Storage / SAN Compatibility Guide For ESX Server 3.5 and ESX Server 3i
by VMware on 01/30/2008 @ http://www.vmware.com/pdf/vi35_san_guide.pdf

VMware ESX Server 3.5 has been tested and deployed in a variety of storage area network (SAN) environments. This guide describes the combination of HBAs (host bus adapters) and storage devices currently tested by VMware and its storage partners.

I/O Compatibility Guide For ESX Server 3.5 and ESX Server 3i
by VMware on 01/30/2008 @ http://www.vmware.com/pdf/vi35_io_guide.pdf

VMware ESX Server Software delivers high performance I/O for PCI-based SCSI, RAID, Fibre Channel, and Ethernet controllers. To achieve high performance, these devices are accessed directly through device drivers in the ESX Server host, and not through a host operating system as with VMware Workstation and GSX Server products.

VMware certifies that specific systems and components are compatible with ESX Server software. Through the VMware Preferred Hardware Partner Program, ESX Server software works with leading server vendors to ensure that appropriate configurations of their current and future server products are certified.

Backup Software Compatibility For ESX Server 3.5 and ESX Server 3i
by VMware on 01/30/2008 @ http://www.vmware.com/pdf/vi35_backup_guide.pdf

This is not an exhaustive list of all backup software packages and versions that are compatible with ESX Server software, but it represents those servers that VMware or its partners have tested with the current release of ESX Server software.

Systems Compatibility Guide For ESX Server 3.5 and ESX Server 3i
by VMware on 01/30/2008 @ http://www.vmware.com/pdf/vi35_systems_guide.pdf

VMware ESX Server is tested for compatibility with a variety of major guest operating systems running in virtual machines. Additionally, VMware ESX Server is tested for compatibility with currently shipping platforms from the major server makers in pre-release testing. Our goal is to support a variety of storage and network adapters used as standard options for these platforms as they come to market.

SPECweb2005 Performance
by VMware on 01/28/2008 @ http://www.vmware.com/files/pdf/specweb_perf_final.pdf

Virtualization is revolutionizing data center computing by making it easy for people to run multiple operating systems and multiple applications seamlessly on the same computer. More and more organizations are adopting VMware® Infrastructure 3 for server consolidation and to reduce the total cost of ownership.

VMware ESX Server 3.5 is designed for high performance. With a number of optimizations for superior performance, even the most I/O‐intensive applications perform well when deployed on VMware Infrastructure 3. In this paper we compare the performance of a virtual machine to that of a similarly configured native machine using the industry standard SPECweb2005 workload. In our virtualized tests we achieved close to 85 percent of native throughput performance using the highly network‐intensive SPECweb2005 workload. In the tests focused on measuring latency, we did not observe any noticeable difference in application latency between the native and virtual environments. These results demonstrate that users need not sacrifice performance in order to embrace the benefits of virtualization technology.

Using the Wyse V10L and S10 Thin Clients with VDM
by VMware on 01/28/2008 @ http://www.vmware.com/files/pdf/VDM_Wyse_tech_note.pdf

This technical note provides a reference for configuring Wyse V10L and S10 devices for use with VDM. VDM supports the Wyse V10L and S10 Thin Client devices, running Wyse Thin OS version 5.3.0.9 or later. For detailed information about deploying and managing Wyse thin client devices, contact Wyse directly.

Wyse Thin Client devices with versions of the Wyse Thin OS earlier than 5.3.0.9 do not support HTTPS connection to a VDM Server. You must upgrade the Wyse Thin OS to version 5.3.0.9 or later to use HTTPS.

Unlike VDM Web Access or VDM Client, Wyse thin clients connect to desktop virtual machines directly using RDP and connections do not pass through a VDM Connection Server. As a result Wyse thin clients cannot be used in DMZ deployments.

Windows XP Deployment Guide
by VMware on 01/27/2008 @ http://www.vmware.com/files/pdf/XP_guide_vdi.pdf

VMware® Virtual Desktop Infrastructure (VDI) transforms the way customers use and manage desktop operating systems. Desktop instances can be deployed rapidly in secure data centers to facilitate high availability and disaster recovery, protect the integrity of enterprise information, and remove data from local devices that are susceptible to theft or loss. Isolating each desktop instance in its own virtual machine eliminates typical application compatibility issues and improves users’ personal computing environments.

This guide offers best practices for creating Windows XP-based templates for VMware VDI-based solutions and for preparing the same templates for use with Virtual Desktop Manager 2.

VMware VDM 2 Load Balancing Guide
by VMware on 01/27/2008 @ http://www.vmware.com/files/pdf/VDM_2_lb_guide.pdf

Once appropriate server-grade hardware has been selected, load balancing become an important consideration for addressing a configuration’s scalability and fault tolerance.

In general, load-balanced configurations use multiple VDM Connection Servers installed in a primary-and-replica manner, with the first server installed as the primary and subsequent servers are installed as replicas. VDM Connection Servers provide session management and handle all incoming client requests, directing them to the appropriate virtual desktop session, and VDM Security Servers provide SSL tunneling capabilities for encrypting communication between the client devices and the VDM Connection Servers.

The configuration of a load-balanced solution largely depends on the requirements of the organization for which it is being deployed. Companies that already have a load balancing solution in place may be able to utilize it for VDI since the load generated by the VDI solution is minimal. Both hardware-based load balancing appliances and inexpensive (or free) software-based load balancing products can be considered as candidate solutions.

Using VMware VDI and vmSight for Stronger and Sustainable HIPAA and PCI Compliance
by VMware on 01/27/2008 @ http://www.vmware.com/files/pdf/HIPAA_PCI_vdi_compliance.pdf

For many organizations, HIPAA and PCI compliance is no longer optional. Penalties are increasingly stiff, and the ultimate cost of non-compliance – the breach of personal data – can damage organizations in multiple ways. Compliance, and validation of compliance, is not easily achieved. It is especially difficult when computing environments are widely distributed and not all computers are centrally managed.

Virtualization technologies, including virtual desktops, offer an improved means to centralize computing, management and monitoring while still providing users local access and full functionality. For organizations seeking stronger HIPAA and PCI compliance, relying on virtual desktops for access to sensitive systems provides both cost savings and increased manageability and security. With an integrated compliance monitoring and reporting system such as the vmSight virtual network intelligence suite, virtual desktops can be used to increase compliance and reduce data breaches while also reducing IT costs.

Comparison of Storage Protocol Performance
by VMware on 01/24/2008 @ http://www.vmware.com/files/pdf/storage_protocol_perf.pdf

This paper demonstrates that the four network storage connection options available to ESX Server are all capable of reaching a level of performance limited only by the media and storage devices. And even with multiple virtual machines running concurrently on the same ESX Server host, the high performance is maintained. The data on CPU costs indicates that Fibre Channel and hardware iSCSI are the most CPU efficient, but in cases in which CPU consumption is not a concern, software iSCSI and NFS can also be part of a high‐performance solution.

VirtualCenter Database Maintenance: SQL Server
by VMware on 01/23/2008 @ http://www.vmware.com/files/pdf/vc_microsoft_sql_server.pdf

This document discusses ways to maintain the VirtualCenter database for increased performance and manageability. It covers all versions of VirtualCenter 2.0. The recommendations here pertain specifically to SQL Server, version 2000 and later. This document assumes the reader has a working knowledge of SQL Server.

What Is New for Storage in Virtual Infrastructure 3 Release 3.5
by VMware on 01/06/2008 @ http://www.vmware.com/files/pdf/new_storage_features_3_5_v6.pdf

With the launch of VMware Infrastructure 3 version 3.5 comes a new version of VMware’s hypervisor, ESX Server 3.5. One of the core focus areas for Virtual Infrastructure 3.5 is to enable greater uptime and agility in the management of the virtual infrastructure environment. VMware has addressed this by bringing new levels of flexibility and connectivity options to Virtual Infrastructure 3 version 3.5.

VMware Consolidated Backup: Improvements in Version 3.5
by VMware on 12/28/2007 @ http://www.vmware.com/files/pdf/vcb_35_new.pdf

This paper descibes the new features of VCB introduced with VI 3, version 3.5. It explains the features and how customers can benefit with those features.

VMware VMotion and CPU Compatibility
by VMware on 12/27/2007 @ http://www.vmware.com/files/pdf/vmotion_info_guide.pdf

This white paper focuses exclusively on CPU compatibility and outlines CPU compatibility checks performed by VMware VirtualCenter before allowing migration with VMotion. It describes why some CPU compatibility constraints make VMotion possible only between certain revisions of CPUs. The appendices detail some differences in features and extensions in current CPUs and describe procedures that could be used to relax some CPU compatibility constraints to facilitate VMotion.

VMware Consolidated Backup
by VMware on 12/27/2007 @ http://www.vmware.com/files/pdf/vcb_best_practices.pdf

This whitepaper is intended to enhance the readerâ€™s knowledge by discussing deployment considerations, best practices recommendations, and troubleshooting tips when implementing VMware Consolidated Backup. The whitepaper is a collection of information gathered by real world field implementations, expertise of VMware Professional Services and in-house testing. Insightful information useful in troubleshooting VCB environment is also provided.

Performance Characteristics of VMFS and RDM
by VMware on 12/07/2007 @ http://www.vmware.com/files/pdf/vmfs_rdm_perf.pdf

VMware ESX Server offers three choices for managing disk access in a virtual machine—VMware Virtual Machine File System (VMFS), virtual raw device mapping (RDM), and physical raw device mapping. It is very important to understand the I/O characteristics of these disk access management systems in order to choose the right access type for a particular application. Choosing the right disk access management method can be a key factor in achieving high system performance for enterprise‐class applications.

This study provides performance characterization of the various disk access management methods supported by VMware ESX Server. The goal is to provide data on performance and system resource utilization at various load levels for different types of work loads. This information offers you an idea of relative throughput, I/O rate, and CPU efficiency for each of the options so you can select the appropriate disk access method for your application.

Firewall Configurations for Backup Clients on ESX Server 3
by VMware on 12/09/2007 @ http://www.vmware.com/pdf/vi3_35_25_vcb_firewalls.pdf

This document provides information about how to configure connections between different backup software products and VMware® ESX Server 3 hosts. The document explains how to open predefined firewall ports for supported backup products and how to open specific ports from a command line. If the backup product you use requires additional configuration changes to work with ESX Server 3, the document describes any specific steps you need to perform.

Virtual Machine Failure Monitoring
by VMware on 12/09/2007 @ http://www.vmware.com/pdf/vi3_35_25_vmha.pdf

VMware® High Availability (VMware HA) monitors your virtual infrastructure for ESX Server host failures and restarts virtual machines that are interrupted by those failures on alternate hosts. Starting with ESX Server 3.5, VMware HA can also detect and handle the failures of individual virtual machines and respond appropriately based on your specifications.

With this additional functionality, called Virtual Machine Failure Monitoring, VMware HA deals with both host operating system and guest operating system failures.

Round-Robin Load Balancing
by VMware on 12/09/2007 @ http://www.vmware.com/pdf/vi3_35_25_roundrobin.pdf

VMware® ESX Server 3.5 and ESX Server 3i version 3.5 enhance ESX Server native multipathing by providing experimental support for round‐robin load balancing. This technical note explains how round‐robin load balancing works and how to set it.

Configuring and Troubleshooting N-Port ID Virtualization
by VMware on 12/09/2007 @ http://www.vmware.com/pdf/vi3_35_25_npiv_config.pdf

N‐Port ID Virtualization (NPIV) is an ANSI T11 standard that describes how a single Fibre Channel HBA port can register with the fabric using several worldwide port names (WWPNs). This allows a fabric‐attached N‐port to claim multiple fabric addresses. Each address appears as a unique entity on the Fibre Channel fabric.

The primary source of information on configuring NPIV in a VMware Infrastructure 3 environment is the Fibre Channel SAN Configuration Guide, available on the VMware Web site.

This technical note provides additional details about certain specific NPIV configurations and information on diagnostic techniques that may be helpful as you configure NPIV. It also includes information to help you understand error messages that may appear as you are working with NPIV in a VMware Infrastructure environment.

Enabling NetFlow on Virtual Switches
by VMware on 12/09/2007 @ http://www.vmware.com/pdf/vi3_35_25_netflow.pdf

NetFlow is a general networking tool with multiple uses, including network monitoring and profiling, billing, intrusion detection and prevention, networking forensics, and SOX compliance. NetFlow sends aggregated networking flow data to a third‐party collector (an appliance or server). The collector and analyzer report on various information such as the current top flows consuming the most bandwidth in a particular virtual switch, which IP addresses are behaving irregularly, and the number of bytes a particular virtual machine has sent and received in the past 24 hours.

VMware OVF Tool
by VMware on 12/10/2007 @ http://www.vmware.com/pdf/ovf_tool.pdf

The Open Virtual Machine Format (OVF) is a virtual machine distribution format that supports sharing virtual machines between products and organizations. The format facilitates the use of virtual appliances, which are preconfigured virtual machines that package applications with the operating system they require. Because OVF runs on multiple platforms, a virtual appliance is ready to run without significant additional configuration.

Using VMware ESX Server With Hitachi Data Systems NSC or USP Storage
by VMware on 10/02/2007 @ http://www.vmware.com/pdf/hds_svd_technote.pdf

This technical note discusses using ESX Server hosts with a Hitachi Data Systems (HDS) NSC or USP SAN storage array using FibreChannel (FC) connections. These storage arrays use advanced storage virtualization technology, and setup differs from setup of a nonâ€virtualized SAN storage array that is to be used with ESX Server hosts.

Best Practices for Building Virtual Appliances
by VMware on 11/15/2007 @ http://www.vmware.com/files/pdf/Best_Practices_Building_Virtual_Appliances.pdf

A virtual appliance is a pre-installed, pre-configured operating system and software solution delivered inside a virtual machine. Deploying a software solution as a virtual appliance enables you to build a complete turnkey package that customers are able to download and immediately deploy. Thus, customers skip the time-consuming and often support-intensive task of installing and configuring the appliance. This lets customers focus all their energies on trying or using your solution rather than struggling to get it to run. This document describes the best practices for building a virtual appliance. It covers high level design principles as well as low level details for building virtual appliances ready for certification under the VMware Certified Virtual Appliance program. In turn, virtual appliances built according to these standards will allow your customers or prospective customers to test or use your virtual appliance with all the VMware virtualization platforms.

Management of VMware ESX Server 3i
by VMware on 11/12/2007 @ http://www.vmware.com/files/pdf/ESXServer3i_management.pdf

VMware ESX Server 3i is the next-generation hypervisor, offering improved security, increased reliability, and simplified management. In many respects, the functionality of an ESX Server 3i system is the same as for ESX Server 3. However, the architecture of ESX Server 3i points the way to a new management model for a virtualized infrastructure. This paper covers the management of an ESX Server 3i system and lays out the characteristics of this new management model.

Architecture of VMware ESX Server 3i
by VMware on 11/12/2007 @ http://www.vmware.com/files/pdf/ESXServer3i_architecture.pdf

VMware ESX Server 3i is the next-generation hypervisor, providing a new foundation for virtual infrastructure. This innovative architecture operates independently from any general purpose operating system, offering improved security, increased reliability, and simplified management. This paper describes the architecture and operation of ESX Server 3i and discusses the new management model associated with it.

Understanding Full Virtualization, Paravirtualization, and Hardware Assist
by VMware on 11/11/2007 @ http://www.vmware.com/files/pdf/VMware_paravirtualization.pdf

In 1998, VMware figured out how to virtualize the x86 platform, once thought to be impossible, and created the market for x86 virtualization. The solution was a combination of binary translation and direct execution on the processor that allowed multiple guest OSes to run in full isolation on the same computer with readily affordable virtualization overhead.

The savings that tens of thousands of companies have generated from the deployment of this technology is further driving the rapid adoption of virtualized computing from the desktop to the data center. As new vendors enter the space and attempt to differentiate their products, many are creating confusion with their marketing claims and terminology. For example, while hardware assist is a valuable technique that will mature and expand the envelope of workloads that can be virtualized, paravirtualization is not an entirely new technology that offers an “order of magnitude” greater performance.

While this is a complex and rapidly evolving space, the technologies employed can be readily explained to help companies understand their options and choose a path forward. This white paper attempts to clarify the various techniques used to virtualize x86 hardware, the strengths and weaknesses of each, and VMware’s community approach to develop and employ the most effective of the emerging virtualization techniques. Figure 1 provides a summary timeline of x86

virtualization technologies from VMware’s binary translation to the recent application of kernel paravirtualization and hardware-assisted virtualization.

SQL Server Performance in a VMware Infrastructure 3 Environment
by VMware on 11/11/2007 @ http://www.vmware.com/files/pdf/SQLServerWorkloads.pdf

This performance study clearly demonstrates that VMware Infrastructure 3 provides an excellent production-ready virtualization platform for customers looking to deploy Microsoft SQL Server inside virtual machines. Furthermore, together with virtualization-based distributed infrastructure services such as VMotion, VMware High Availability, and VMware Distributed Resource Scheduler, VMware Infrastructure 3 can provide increased serviceability, efficiency, and reliability for your SQL Server deployments. This should offer transformative cost savings to your dynamic data center.

iSCSI Design Considerations and Deployment Guide
by VMware on 11/05/2007 @ http://www.vmware.com/files/pdf/iSCSI_design_deploy.pdf

With the release of VMware Infrastructure 3, VMware added ESX Server support for iSCSI storage. With rapidly increasing adoption rates, many VMware customers requested iSCSI as an option for storage virtualization and are now deploying it as a lower cost alternative to Fibre Channel SANs. This paper is intended to help you understand the design considerations and deployment options for deploying VMware Infrastructure 3 using iSCSI storage. The first section provides an overview of iSCSI terminology, benefits, and limitations. The second section provides a high-level overview of the VMware iSCSI implementation using either a software initiator or a hardware initiator. The third section provides a detailed set of deployment steps covering both software and hardware initiator options. The paper concludes with two appendices that provide software versus hardware initiator iSCSI performance test results and details on command line options for managing iSCSI from the ESX Server host. This paper highlights trade-offs and factors to consider when deploying iSCSI storage to support VMware Infrastructure 3. It is a complement to, not a replacement for, VMware product documentation.

The Roadmap to Virtual Infrastructure: Practical Implementation Strategies
by VMware on 09/20/2007 @ http://download3.vmware.com/elq/pdf/wp_roadmaptovirtualinfrastructure.pdf

The paper provides recommendations to provide IT management with the most salient best practices and implementation strategies to get started and to accelerate a successful roll out of virtualization technology. These guidelines are based on experiences and best practices accumulated by many of our leading customers and partners.

OVF Open Virtual Machine Format Specification
by VMware on 09/07/2007 @ http://www.vmware.com/pdf/ovf_spec_draft.pdf

The Open Virtual Machine Format (OVF) describes an open, secure, portable, efficient and extensible format for the packaging and distribution of (collections of) virtual machines.

The Open Virtual Machine Format Whitepaper for OVF Specification
by VMware on 09/07/2007 @ http://www.vmware.com/pdf/ovf_whitepaper_specification.pdf

The Open Virtual Machine Format (OVF) specification is a hypervisor-neutral, efficient, extensible, and open specification for the packaging and distribution of virtual appliances composed of one or more VMs. It aims to facilitate the automated, secure management not only of virtual machines but the appliance as a functional unit. For the OVF format to succeed it must be developed and endorsed by ISVs, virtual appliance vendors, operating system vendors, as well as virtual platform vendors, and must be developed within a standards-based framework.

VMware Infrastructure and CA XOsoft's WANSyncHA: Building a DR System
by VMware on 08/19/2007 @ http://www.vmware.com/files/pdf/vi_xosoft.pdf

You can combine two very powerful solutions, VMware Infrastructure and CA XOsoft's WANSyncHA, to provide a multi-layered disaster recovery solution that covers a wide array of contingencies, is extremely cost-effective, and provides an unusually high degree of robustness and simplicity.

Best Practices for Setting Up VMware ACE 2.0 Enterprise Edition
by VMware on 08/20/2007 @ http://www.vmware.com/pdf/ace_best_practices.pdf

VMware ACE 2.0 Enterprise Edition enables you to apply corporate IT policies to a virtual machine containing an operating system, enterprise applications, and data to create a secure, isolated PC environment known as an “ACE virtual machine.” This technical note explains how to set up VMware Workstation with ACE option pack to most efficiently create and manage ACE masters.

Pocket ACE Guide
by VMware on 08/18/2007 @ http://www.vmware.com/pdf/pocket_ace_technote.pdf

The purpose of this technical note is to provide an in-depth view of Pocket ACE, a new feature of VMware ACE 2 Enterprise Edition. This guide is intended for architects or administrators looking for more information about Pocket ACE and how it might be used in their environments. This guide covers the steps required to create an ACE master, policy, and package for deploying Pocket ACE instances.

Disaster Recovery Virtualization: Protecting Production Systems Using VMware Virtual Infrastructure and Double-Take
by VMware on 08/17/2007 @ http://www.vmware.com/files/pdf/DR_VMware_DoubleTake.pdf

This paper describes the use of VMware virtualization and NSI Doubletake replication technology to implement a DR solution. The paper educates the reader about using the virtualization technology to protect physical environment in primary data centers and also talks about benefits of P2V and V2V DR.

Optimized Backup and Recovery for VMware Infrastructure with EMC Avamar
by VMware on 08/13/2007 @ http://www.vmware.com/files/pdf/VMware_and_Avamar_Backup.pdf

This document provides technical details for the different methods of backup and recovery of VMware Infrastructure using EMC Avamar software. It gives a brief overview of VMware Infrastructure and EMC Avamar, outlines the architectural requirements for each method, and reviews advantages and key considerations to consider.

VMware Virtual Networking Concepts
by VMware on 07/29/2007 @ http://www.vmware.com/files/pdf/virtual_networking_concepts.pdf

VMware Infrastructure 3 provides a rich set of networking capabilities that integrate well with sophisticated enterprise networks. With virtual networking, you can network virtual machines in the same way that you do physical machines and can build complex networks within a single ESX Server host or across multiple ESX Server hosts, for production deployments or development and testing purposes. This guide is for VMware Infrastructure 3 users who want a clearer understanding of the basic design of the networking capabilities in VMware Infrastructure 3 and how that design affects deployment in the datacenter.

VMware Virtual Machine File System: Technical Overview and Best Practices
by VMware on 07/25/2007 @ http://www.vmware.com/pdf/vmfs-best-practices-wp.pdf

The VMware Virtual Machine File System (VMFS) is a high performance cluster file system that allows virtualization to scale beyond the boundaries of a single system. This paper gives a technology overview of VMFS, including a discussion of features and their benefits. The paper highlights how VMFS capabilities enable greater scalability and decreased management overhead. It also provides best practices and architectural considerations for deployment of VMFS.

Layers of Virtual Storage in VMware VI3: Configuration without Confusion
by VMware on 07/25/2007 @ http://www.vmware.com/pdf/storage-layers-wp.pdf

Provisioning storage for a virtual environment requires navigating several layers of abstraction, making it desirable to view this as a stack in order to follow the complete path successfully. The purpose of this paper is to define the options, choices and terms that can help trace the route through this complex set of dependencies.

Microsoft Exchange Server 2003 Performance on VMware® ESX Server 3
by VMware on 07/09/2007 @ http://www.vmware.com/pdf/Virtualizing_Exchange2003.pdf

This paper discusses the performance and scalability of Microsoft Exchange Server 2003 when deployed within virtual machines running under VMware ESX Server 3.01.

Using MSCS to Cluster VirtualCenter
by VMware on 06/21/2007 @ http://www.vmware.com/pdf/VC_MSCS.pdf

This document describes the use of Microsoft Cluster Server (MSCS) to provide high availability for VirtualCenter. This capability is available in VirtualCenter 2.x starting with VirtualCenter 2.0.1 Patch 2. You can implement the clustering architecture described in this paper whether VirtualCenter runs on a physical server or in a virtual machine.

Networking Performance in Multiple VMs
by VMware on 06/05/2007 @ http://www.vmware.com/pdf/Multi-VM_Network_Performance.pdf

Scaling and performance of VMware ESX Server 3.0.1 and XenEnterprise 3.2.0 are measured by running a heavy networking load simultaneously in each of several uniprocessor virtual machines. The results are also compared with a single SMP native machine running the same total load.

Using a Network Appliance SAN with VMware Infrastructure 3 to Facilitate Server and Storage Consolidation
by Network Appliance on 05/30/2007 @ http://www.vmware.com/pdf/vmware_netapp_consolidation_wp7018.pdf

This paper describes the real-world experiences of a customer that reduced its x86 server count by 60% and decreased overall operating costs.

Performance Tuning and Benchmarking Guidelines for VMware Workstation 6
by VMware on 05/08/2007 @ http://www.vmware.com/pdf/WS6_Performance_Tuning_and_Benchmarking.pdf

This book provides guidance in obtaining the best possible performance using VMware Workstation 6.0, both in a production environment and when running benchmarks.

Multi-NIC Networking Performance in ESX 3.0.1 and XenEnterprise 3.2.0
by VMware on 04/30/2007 @ http://www.vmware.com/pdf/Multi-NIC_Performance.pdf

In this paper, we characterize the virtualization overheads of virtual machines by measuring the performance under heavy networking in a uniprocessor virtual machine.

Managing VMware VirtualCenter Roles and Permissions
by VMware on 04/16/2007 @ http://www.vmware.com/pdf/vi3_vc_roles.pdf

This paper introduces you to the way Virtual Infrastructure 3 controls access to resources and describes techniques you can use to assign appropriate access rights efficiently. It explains the concept of roles, provides information to help in the design of custom roles, and gives recommendations for how to work with roles and privileges in VirtualCenter.

Running VirtualCenter in a Virtual Machine
by VMware on 04/03/2007 @ http://www.vmware.com/pdf/vi3_vc_in_vm.pdf

Running VirtualCenter in a virtual machine is fully supported by VMware to the same degree as if it were installed on a physical server. The purpose of this paper is to provide guidelines on how to deploy VirtualCenter in a virtual machine, including sizing, installation, functionality, and configuration of VMware HA.

SAN System Design and Deployment Guide
by VMware on 03/19/2007 @ http://www.vmware.com/pdf/vi3_san_design_deploy.pdf

This guide, or “cookbook,” describes how to design and deploy virtual infrastructure systems using VMware Infrastructure 3 with SANs (storage area networks). It describes SAN options supported with VMware Infrastructure 3 and also describes benefits, implications, and disadvantages of various design choices.

VirtualCenter Monitoring and Performance Statistics
by VMware on 03/19/2007 @ http://www.vmware.com/pdf/vi3_monitoring_statistics_note.pdf

This document lists the metrics available for monitoring from VirtualCenter. It includes guidelines for setting the Statistics Collection Level, and indicates the types of metrics available for each level.

VirtualCenter 2.x: Using Database Views
by VMware on 02/14/2007 @ http://www.vmware.com/pdf/vc_dbviews_2x.pdf

This document describes the relationships of the provided database views in your VMware VirtualCenter database. This document does not describe how to create or use database views. This document describes the database views for VirtualCenter version 2.x.

Security Design of the VMware Infrastructure 3 Architecture
by VMware on 02/22/2007 @ http://www.vmware.com/pdf/vi3_security_architecture_wp.pdf

VMware Infrastructure 3 is one of the most secure and robust virtualization platforms available. VMware has both the technology and the processes to ensure that this high standard is maintained in all current and future products. This document discusses the architecture of VMware Infrastructure 3, focusing on the security aspects of the design.

VMware Infrastructure 3 Security Hardening
by VMware on 02/21/2007 @ http://www.vmware.com/pdf/vi3_security_hardening_wp.pdf

This paper provides recommendations for steps you can take to ensure that your VMware Infrastructure 3 environment is properly secured. It also explains in detail the security-related configuration options of the components of VMware Infrastructure 3 and the consequences for security of enabling certain capabilities.

A Performance Comparison of Hypervisors
by VMware on 01/31/2007 @ http://www.vmware.com/pdf/hypervisor_performance.pdf

This paper provides a quantitative and qualitative comparison of two virtualization hypervisors available for the x86 architecture — VMware ESX Server 3.0.1 and open-source Xen 3.0.3 — to validate their readiness for enterprise datacenters.

Performance Tuning Best Practices for ESX Server 3
by VMware on 01/25/2007 @ http://www.vmware.com/pdf/vi_performance_tuning.pdf

The paper provides a list of performance tips that cover the most performance-critical areas of Virtual Infrastructure 3 (VI3). The intended audience is system administrators who have already deployed VI3 and are looking to maximize their performance.

Deploying Microsoft Exchange in VMware Infrastructure
by RapidApp on 01/18/2007 @ http://www.vmware.com/pdf/exchange_best_practices.pdf

Deploying Microsoft Exchange in VMware Infrastructure offers many advantages, including higher availability,simplified recovery, and more efficient use of resources. This paper discusses best practices for running Exchange 2003 in a VMware Infrastructure environment, in the areas of servers, virtual machines, storage, and implementation. In addition, a case study shows an example architecture, and the Appendix provides some sample metrics for Exchange performance.

Replacing VirtualCenter Server Certificates
by VMware on 12/19/2006 @ http://www.vmware.com/pdf/vi_vcserver_certificates.pdf

This Technical Note provides information about replacing the default certificates supplied with VirtualCenter Server hosts.

VDI -- A New Desktop Strategy
by Foedus on 12/05/2006 @ http://www.vmware.com/pdf/vdi_strategy.pdf

VDI allows IT administrators to host and administer user desktops on Virtual Infrastructure in the datacenter. Users access their desktop using a remote desktop protocol. While sharing similarities with other computing models, VDI offers many new and compelling benefits for increasing manageability, performance, and security of user desktops/PCs. This paper compares VDI to other user management strategies and highlights VDI’s benefits for particular use cases.

Configuring Management Agents for ESX Server 3.0.1
by VMware on 10/05/2006 @ http://www.vmware.com/pdf/esx30_cfg_mgmt_tools.pdf

This technical note describes how to configure the selected management agents for use with ESX Server 3.0.1: Dell OpenManage 5.1 agents, HP Insight Manager 7.6.0 agents, and IBM Director 5.10.3 agents.

Scaling IBM DB2 in VMware Infrastructure 3 Environment
by VMware on 10/09/2006 @ http://www.vmware.com/pdf/db2_scalability_wp_vi3.pdf

This white paper characterizes the performance of a web-based OLTP workload using IBM DB2 9 and running in VMware Infrastructure 3 environment. The results described in this paper support the conclusion that running IBM DB2 in VMware virtual machines can provide an effective production ready platform for hosting multiple virtualized OLTP workloads.

Improving Scalability for Citrix Presentation Server
by VMware on 10/22/2006 @ http://www.vmware.com/pdf/esx_citrix_scalability.pdf

This white paper presents data showing the number of Citrix sessions you can sustain when running Citrix Presentation Server 4.0 inside a virtual machine, and then measures scalability with 1, 2, 4 and 8 VMs. For ESX 3, engineering improved Citrix performance in a virtual machine and we want to showcase those gains, as well as present best practices for performance.

Ready Time Observations
by VMware on 10/31/2006 @ http://www.vmware.com/pdf/esx3_ready_time.pdf

To achieve best performance in a consolidated environment, you must consider ready time — the time a virtual machine must wait in a ready-to-run state before it can be scheduled on a CPU. This paper provides information to help you understand the factors that influence ready time on an ESX Server 3.0 system.

Using VMware Infrastructure for Backup and Restore
by VMware on 10/26/2006 @ http://www.vmware.com/pdf/esx3_backup_wp.pdf

This white paper describes what resources should be backed up on an ESX Server system and explains the options available for that backup, including advantages and disadvantages of each option. The paper also offers some recommendations based on these trade-offs.

This paper applies to VMware Infrastructure 3. The corresponding paper for ESX Server 2.x is available at http://www.vmware.com/vmtn/resources/237

Recommendations for Aligning VMFS Partitions
by VMware on 08/01/2007 @ http://www.vmware.com/pdf/esx3_partition_align.pdf

Partition alignment is a known issue in physical file systems, and its remedy is well-documented. The goal of the testing reported in this paper was to validate the assumption that unaligned partitions also impose a penalty when the partition is a VMware Virtual Machine File System (VMware VMFS) partition.

This paper lists a summary of the results of our testing, recommendations for VMware VMFS partition alignment, and the steps needed to create aligned VMware VMFS partitions.

The Role of Memory in VMware ESX Server 3
by Kingston on 10/23/2006 @ http://www.vmware.com/pdf/esx3_memory.pdf

The memory management capabilities of VMware ESX Server 3.0 provide a unique and sophisticated way to maximize the usage of physical memory within a single box. For many workloads, memory is the limiting factor, and effective memory management enables more virtual machines to share a single server, increasing ROI for consolidation. Advances in virtualization, CPU, and memory technology make the addition of memory one of the most effective investments for maximizing the utilization of an ESX Server host.

Patch Management for ESX Server 3
by VMware on 09/06/2007 @ http://www.vmware.com/pdf/esx3_esxupdate.pdf

The VMware ESX 3 Server software maintenance tool, esxupdate, is a utility that efficiently updates VMware ESX Server hosts. Use it to install software updates, enforce software update policies, and track installed software. The benefits of this mechanism include smaller distribution downloads, atomic updates, ability to automate update deployment, selectability by update classification, and automatic dependency resolution.

SAN Conceptual and Design Basics
by VMware on 10/03/2006 @ http://www.vmware.com/pdf/esx_san_cfg_technote.pdf

To use ESX Server effectively with a SAN, youre expected to be familiar with the SAN technology. This white paper offers a brief introduction to some basic SAN concepts, but doesn't aim to be an exhaustive source of information on SANs.

Enabling Active Directory Authentication with ESX Server
by VMware on 10/02/2006 @ http://www.vmware.com/pdf/esx3_esxcfg_auth_tn.pdf

This document provides information about how to configure ESX Server to use Active Directory for authentication.

This paper applies to VMware Infrastructure 3. The corresponding paper for ESX Server 2.x is available at http://www.vmware.com/vmtn/resources/429

The Portable Datacenter
by VMware on 09/25/2006 @ http://www.vmware.com/pdf/portable_datacenter.pdf

Using VMware technology as a foundation, Smartronix, a global professional solutions provider,created a complete datacenter that is self-contained and easily transported to any location at a moment’s notice. This paper documents the architecture of this portable datacenter solution.

VMmark: A Scalable Benchmark for Virtualized Systems
by VMware on 09/24/2006 @ http://www.vmware.com/pdf/vmmark_intro.pdf

This paper presents VMmark, a novel benchmark for quantifying the performance of virtualized environments. VMmark is designed as a tile-based benchmark consisting of a diverse set of workloads commonly found in the datacenter. The workloads comprising each tile are run simultaneously in separate virtual machines at load levels that are typical of virtualized environments. The performance of each workload is measured and then combined with the other workloads to form the score for the individual tile. Multiple tiles can be run simultaneously to increase the overall score.

VMware ESX Server Using EMC Symmetrix Storage Systems
by EMC on 09/19/2006 @ http://www.emc.com/techlib/pdf/300-003-507_a01_elccnt_0.pdf

This solutions guide describes integration of the VMware ESX Server with EMC Symmetrix networked storage systems—and software including EMC Symmetrix Remote Data Facility (SRDF) and EMC TimeFinder local replication software. Topics covered include installation, setup, configuration, and best practices; how TimeFinder can be used with VMware ESX; and how SRDF can be leveraged with VMware ESX Server to provide business restart processing.

Double-Take Replication in the VMware Environment
by VMware on 09/18/2006 @ http://www.vmware.com/pdf/vmware_doubletake.pdf

This technical white paper talks about the way Double-Take's replication solution can be used in VMware environment. It also talks about a customer experience and lists a few example scenarios.

Using VMware ESX Server with 3PAR Utility Storage to Optimize Virtual Computing Deployments
by VMware on 10/15/2006 @ http://www.vmware.com/pdf/esx3_3par_util_stor.pdf

This whitepaper discusses the cost of over-provisioning servers and storage in an IT infrastructure, and how this unnecessary cost can be reduced and even eliminated through combining server virtualization technology and utility storage. The combination of VMware ESX Server and 3PAR Utility Storage results in a powerful virtualized utility computing platform.

VMware ESX Server Using EMC CLARiiON Storage Systems
by EMC on 08/28/2006 @ http://www.emc.com/techlib/pdf/H2197_VMWare_ESX_CLARiiON_Stor_Syst_ldv.pdf

This solutions guide describes integration of the VMware ESX Server with EMC CLARiiON networked storage systems—and software including EMC SnapView, EMC MirrorView, and EMC SAN Copy. Download this guide for detailed information and instruction on how CLARiiON storage systems and software enhance VMware functionality by providing capabilities to efficiently grow, clone, and/or remotely replicate virtual environments.

Planning an Upgrade to VMware Infrastructure 3
by VMware on 10/15/2006 @ http://www.vmware.com/pdf/vi3_upgrade_plan.pdf

Because of the breadth of the changes in VMware Infrastructure 3, environments that currently run ESX Server 2.x and VirtualCenter 1.x must be upgraded with procedures that utilize special capabilities designed into the product. This guide aims to help IT administrators and architects understand the upgrade process and plan for it appropriately.

VDI Server Sizing and Scaling
by VMware on 08/24/2006 @ http://www.vmware.com/pdf/vdi_sizing_vi3.pdf

VMware Virtual Desktop Infrastructure (VDI) is a server-based computing offering that provides desktop environments as an enterprise hosted service. VDI allows administrators to maintain and manage all user applications, data, and environments in the centrally located data center.

This white paper describes the testing methodology, results, and analysis and sizing guidelines for setting up Virtual Desktop Infrastructure using VMware Infrastructure 3.

Fast Transparent Migration for Virtual Machines
by VMware on 04/09/2005 @ http://www.vmware.com/pdf/usenix_vmotion.pdf

This paper describes the design and implementation of a system that uses virtual machine technology to provide fast, transparent application migration. This is the first system that can migrate unmodified applications on unmodified mainstream Intel x86-based operating system, including Microsoft Windows, Linux, Novell NetWare and others. Neither the application nor any clients communicating with the application can tell that the application has been migrated. Experimental measurements show that for a variety of workloads, application downtime caused by migration is less than a second.

Memory Resource Management in VMware ESX Server
by VMware on 12/08/2002 @ http://www.vmware.com/pdf/usenix_resource_mgmt.pdf

This paper introduces several novel ESX Server mechanisms and policies for managing memory. A ballooning technique reclaims the pages considered least valuable by the operating system running in a virtual machine. An idle memory tax achieves efficient memory utilization while maintaining performance isolation guarantees. Content-based page sharing

and hot I/O page remapping exploit transparent page remapping to eliminate redundancy and reduce copying overheads. These techniques are combined to efficiently support virtual machine workloads that overcommit memory.

Virtualizing I/O Devices on VMware Workstation’s Hosted Virtual Machine Monitor
by VMware on 06/24/2001 @ http://www.vmware.com/pdf/usenix_io_devices.pdf

This paper focuses on VMwareWorkstation’s approach to virtualizing I/O devices. This paper studies the virtualization and performance of an Ethernet adapter on VMware Workstation. Results indicate that with optimizations, VMware Workstation’s hosted virtualization architecture can match native I/O throughput on standard PCs.

An Analysis of Disk Performance in VMware ESX Server Virtual Machines
by VMware on 10/26/2003 @ http://www.vmware.com/pdf/wwc_performance.pdf

This paper studies the performance of a key component of the ESX Server architecture: its storage subsystem. We characterize the performance of native systems and virtual machines using a series of disk microbenchmarks on several different storage systems. We show that the virtual machines perform well compared to native, and that the I/O behavior of virtual machines closely matches that of the native server. We then discuss how the microbenchmarks can be used to estimate virtual machine performance for disk-intensive applications by studying two workloads: a simple file server and a commercial mail server.

A Comparison of Software and Hardware Techniques for x86 Virtualization
by VMware on 08/08/2006 @ http://www.vmware.com/pdf/asplos235_adams.pdf

The main technical contributions of this paper are (1) a review of VMware Workstation’s software VMM, focusing on performance properties of the virtual instruction execution engine; (2) a review of the emerging hardware support, identifying performance trade-offs; (3) a quantitative performance comparison of a software and a hardware VMM.

Using VMware ESX Server with IBM WebSphere Application Server
by VMware on 08/20/2006 @ http://www.vmware.com/partners/vmware/ESX_WAS_WP.pdf

This paper looks at the best practices when running VMware ESX Server and IBM WebSphere Application Server on a large symmetric multiprocessor (SMP) server to achieve the best application performance (throughput and response time). Our tests concluded that the use of VMware software is an excellent option for those wanting to run WebSphere Application Server in a virtualized SMP environment.

Improving Guest Operating System Accounting for Descheduled Virtual Machines in ESX Server 3.x Systems
by VMware on 08/20/2006 @ http://www.vmware.com/pdf/vi3_esx_vmdesched.pdf

This document explains how to install and monitor VMDesched on Linux and Windows guest operating systems. It also describes timer interrupt virtualization issues resolved by VMDesched and how VMDesched works.

When installed, the VMDesched component provides two key benefits:

• Improved accuracy for guest operating system CPU time accounting when physical CPU resources are overcommitted.

• Improved guest operating system timekeeping with respect to real time.

The experimental VMware Descheduled Time Accounting component, VMDesched, is an optional new component of VMware Tools. VMDesched is available starting with ESX Server 3.0.

VMware ESX Server: Third-Party Software in the Service Console
by VMware on 08/02/2006 @ http://www.vmware.com/pdf/esx_3p_scvcons.pdf

In general, VMware recommends that users avoid running third-party software in the VMware ESX Server service console. Clear exceptions to this policy are software packages explicitly identified in ESX Server compatibility guides. This technical note outlines the reasons for this recommendation and guidelines developers should follow when writing software to run in the ESX Server service console.

Running Microsoft Exchange Server in a Virtual Machine Using ESX Server 2.5
by VMware on 07/21/2006 @ http://www.vmware.com/pdf/exchange_esx25_wp_eng.pdf

This white paper provides lab-tested guidance for implementing an enterprise messaging system using Microsoft Exchange messaging products and VMware ESX Server 2.5.

The paper provides referential and prescriptive guidance to enable a customer or solution provider to adequately plan, build, deploy, and operate an enterprise messaging system that results in the following benefits to an organization:

• Availability to meet the service level agreements (SLAs)

• Security to meet the business requirements

• Scalability to meet the projected business volumes

• Predictable and reliable performance from pre-tested implementations

• Reduced implementation time, cost, and operational risk, leading to a faster time to benefit

Configuring VMware ACE as the Primary Desktop Environment: Sample Registry File
by VMware on 10/06/2005 @ http://download3.vmware.com/software/ace/ace-secure-template.reg.txt

The sample registry file to accompany the technical note Configuring VMware ACE as the Primary Desktop Environment, which explains how to configure VMware ACE and a host computer so the virtual machine running in VMware ACE is the primary desktop environment.

VMware ESX Server Ethernet-based Storage Configuration
by VMware on 06/15/2006 @ http://www.vmware.com/pdf/iscsi_storage_esx.pdf

This technical note examines key issues in configuring iSCSI storage with virtual machines running on an ESX Server system.

HCL: I/O Compatibility Guide for ESX Server 3.x
by VMware on 01/03/2008 @ http://www.vmware.com/pdf/vi3_io_guide.pdf

VMware ESX Server Software delivers high performance I/O for PCI-based SCSI, RAID, Fibre Channel, and Ethernet controllers. To achieve high performance, these devices are accessed directly through device drivers in the ESX Server host, and not through a host operating system as with VMware Workstation and GSX Server products.

VMware certifies that specific systems and components are compatible with ESX Server software. Through the VMware Preferred Hardware Partner Program, ESX Server software works with leading server vendors to ensure that appropriate configurations of their current and future server products are certified.

HCL: Backup Software Compatibility for ESX Server 3.x
by VMware on 12/20/2007 @ http://www.vmware.com/pdf/vi3_backup_guide.pdf

This is not an exhaustive list of all backup software packages and versions that are compatible with ESX Server software, but it represents those servers that VMware or its partners have tested with the current release of ESX Server software.

HCL: Systems Compatibility Guide for ESX Server 3.x
by VMware on 01/03/2008 @ http://www.vmware.com/pdf/vi3_systems_guide.pdf

VMware ESX Server is tested for compatibility with a variety of major guest operating systems running in virtual machines. Additionally, VMware ESX Server is tested for compatibility with currently shipping platforms from the major server makers in pre-release testing. Our goal is to support a variety of storage and network adapters used as standard options for these platforms as they come to market.

HCL: Storage Compatibility Guide for ESX Server 3.x
by VMware on 01/09/2008 @ http://www.vmware.com/pdf/vi3_san_guide.pdf

VMware ESX Server 3.x has been tested and deployed in a variety of storage area network (SAN) environments. This guide describes the combination of HBAs (host bus adapters) and storage devices currently tested by VMware and its storage partners.

Streamlining Patch Testing and Deployment
by VMware on 06/15/2004 @ http://www.vmware.com/pdf/gsx_landesk.pdf

Using VMware GSX Server with LANDesk Management Suite to improve patch deployment speed and reliability.

Using Clones to Streamline Development
by VMware on 06/09/2005 @ http://www.vmware.com/pdf/ws5_clones_technote.pdf

Use Workstation 5, with its linked clones and multiple snapshots features, to set up a repository of clean development environments in a series of virtual machines and then quickly clone and download a desired setup in minutes.

Development in Multitier Configurations with Workstation Teams
by VMware on 06/09/2005 @ http://www.vmware.com/pdf/ws5_teams_technote.pdf

Use Workstation 5, with its teams feature, to set up an entire three-tier configuration for development and testing, virtually, on a single physical machine.

Performance Benchmarking Guidelines for VMware Workstation 5.5
by VMware on 04/14/2006 @ http://www.vmware.com/pdf/WS55_Benchmarking_Guidelines.pdf

This white paper provides guidance in implementing benchmark tests using VMware Workstation 5.5. It addresses benchmarking methodology, configuring for performance, ensuring “apples-to-apples” comparisons, and avoiding common pitfalls.

Building Virtual Infrastructure with VMware VirtualCenter
by VMware on 07/29/2004 @ http://www.vmware.com/pdf/vi_wp.pdf

This paper demonstrates how adopting virtual infrastructure helps organizations drive costs out of their infrastructure, respond faster to business needs and increase consistency of operations.

VirtualCenter Technical Best Practices
by VMware on 07/19/2005 @ http://www.vmware.com/pdf/vc_technical_best.pdf

This presentation provides an overview of VirtualCenter product components, system requirements, the VirtualCenter database, authentication, sample configurations, and VMotion considerations.

Comparing the MUI, VirtualCenter, and vmkusage
by VMware on 03/18/2005 @ http://www.vmware.com/pdf/mui_vmkusage2.pdf

This paper compares three VMware performance tools and highlights where the values of the output can differ, focusing on CPU and memory statistics.

Performance Problem Report Check List
by VMware on 04/08/2005 @ http://www.vmware.com/pdf/esx2_checklist1.0.pdf

This paper presents a check list that VMware Support uses to iteratively eliminate potential performance issues until the system is running optimally.

Using vmkusage to Isolate Performance Problems
by VMware on 03/23/2005 @ http://www.vmware.com/pdf/esx2_using_vmkusage.pdf

This paper introduces vmkusage and describes how to use it to analyze and isolate ESX performance problems.

ESX Performance Tips and Tricks
by VMware on 02/07/2005 @ http://www.vmware.com/pdf/esx_performance_tips_tricks.pdf

This paper provides several useful tips for tuning and troubleshooting your ESX Server machine.

Using esxtop to Troubleshoot Performance Problems
by VMware on 12/23/2004 @ http://www.vmware.com/pdf/esx2_using_esxtop.pdf

This paper describes the process for troubleshooting a suspected performance problem at the virtualization layer (i.e., the ESX Server layer) using the esxtop tool.

Representing Physical Machines in the Virtual World
by VMware on 12/23/2004 @ http://www.vmware.com/pdf/esx2_machine_representation.pdf

This paper describes the process for troubleshooting a suspected performance problem at the machine representation layer (i.e., virtual machine layer).

Isolating Performance Problems
by VMware on 12/23/2004 @ http://www.vmware.com/pdf/esx2_isolating_performance.pdf

This paper provides an overview of the process used to troubleshoot performance problems in the ESX Server environment.

ESX Server Architecture and Performance Implications
by VMware on 08/03/2005 @ http://www.vmware.com/pdf/esx2_performance_implications.pdf

This paper uncovers ESX Server software architecture and provides insights into the techniques used for efficient resource control. It also describes advanced scheduling configuration options and performance tools for measuring system resource utilization.

NIC Bonding and VLANs on IBM BladeCenter
by VMware on 08/10/2004 @ http://www.vmware.com/pdf/esx21_IBM_NIC_VLAN.pdf

This paper covers topics including load balancing, failover only mode, and switch failover mode.

Configuring and Installing HP Blade Servers
by VMware on 04/28/2004 @ http://www.vmware.com/pdf/esx21_HP_blade.pdf

This paper helps you maximize your ESX Server experience on HP blade servers.

Configuring and Installing IBM BladeCenter
by VMware on 03/31/2004 @ http://www.vmware.com/pdf/esx21_IBM_blade.pdf

This paper helps you maximize your ESX Server experience on IBM blade servers.

Login Using Active Directory
by VMware on 03/02/2005 @ http://www.vmware.com/pdf/esx_authentication_AD.pdf

This document explains how to use a Microsoft Windows 2000 Active Directory domain as an authentication source.

This paper applies to ESX Server 2.x. The corresponding paper for VMware Infrastructure 3 is available at http://www.vmware.com/vmtn/resources/582

Login Using Windows NT
by VMware on 05/13/2004 @ http://www.vmware.com/pdf/esx_authentication_NT.pdf

This document explains how to use Microsoft Windows NT 4.0 with domain as an authentication source in the ESX Server 2.1 environment.

Using VERITAS NetBackup DataCenter
by VMware on 03/24/2005 @ http://www.vmware.com/pdf/veritasnetbu.pdf

This technical note describes basic procedures for using VERITAS NetBackup DataCenter version 4.5 to backup VMware ESX Server files.

Using AnywhereUSB to Connect USB Devices
by VMware on 03/01/2006 @ http://www.vmware.com/pdf/esx_anywhereusb2.pdf

AnywhereUSB enables a virtual machine running on VMware ESX Server to send USB data through its standard Ethernet interface over an Ethernet or wireless IP network to nearly any USB device.

Providing LUN Security
by VMware on 03/10/2006 @ http://www.vmware.com/pdf/esx_lun_security.pdf

VMware ESX Server provides strong security and performance isolation for virtual machine storage. This technical note describes the underlying mechanisms and some configuration differences between physical and virtual environments as they relate to SAN storage.

Virtual to Physical Documentation and Sample Configurations
by VMware on 06/18/2004 @ http://www.vmware.com/support/v2p/index.html

The V2P Technical Note details how a customer can convert an existing virtual machine to a physical machine using third party tools.

Converting Image Files into Virtual Machine Disks
by VMware on 07/13/2004 @ http://www.vmware.com/pdf/p2v_thirdpartyimage.pdf

This technical note describes how to create virtual disks, with or without operating systems, using third party tools, compatible with VMware virtual machines.

P2V Assistant Tested Hardware
by VMware on 05/24/2006 @ http://www.vmware.com/pdf/p2v_hardware.pdf

Listing of systems, SCSI / RAID controllers, NICs, and Fiber Channels tested against P2V.

VMware Infrastructure 3: Pricing, Packaging and Licensing Overview
by VMware on 06/02/2006 @ http://www.vmware.com/pdf/vi_pricing.pdf

Simplifying how customers purchase, deploy and get support for VMware data center products, the VMware Infrastructure 3 product suite introduces significant changes to pricing, packaging and licensing for all products included in the suite.

HCL: VMware ESX Server Supported Backup Tools
by VMware on 10/07/2007 @ http://www.vmware.com/support/esx25/doc/backup_tools_links.html

Links to ESX Server 2.5 supported backup clients.

HCL: VMware ESX Server 3.x and 2.5.x Supported System Management Clients
by VMware on 12/18/2007 @ http://www.vmware.com/support/esx25/doc/sys_mgmt_links.html

This document lists the links to ESX Server 3.x and 2.5 supported system management clients.

HCL: Backup Software Compatibility for ESX Server 2.x
by VMware on 10/07/2007 @ http://www.vmware.com/pdf/esx_backup_guide.pdf

This document describes the specific backup packages tested by VMware.

HCL: SAN Compatibility Guide for ESX Server 2.x
by VMware on 11/07/2007 @ http://www.vmware.com/pdf/esx_SAN_guide.pdf

VMware ESX Server has been tested and deployed in a variety of SAN environments. However, in practice, because every customer's device combination, topology, and configuration are unique, VMware recommends that VMware professional services be engaged to install and configure the initial ESX Server installation in your SAN environment.

This paper describes in detail the combinations of HBAs and storage devices currently tested by VMware and its storage partners.

HCL: I/O Compatibility Guide for ESX Server 2.x
by VMware on 11/07/2007 @ http://www.vmware.com/pdf/esx_io_guide.pdf

VMware ESX Server delivers high performance I/O for PCI-based SCSI, RAID, Fibre Channel, and Ethernet controllers. To achieve high performance, these devices are accessed directly through device drivers in the ESX Server machine and not through a host operating system as with VMware Workstation and GSX Server products.

HCL: Systems Compatibility Guide for ESX Server 2.x
by VMware on 12/11/2007 @ http://www.vmware.com/pdf/esx_system_guide.pdf

VMware ESX Server has an architecture that delivers high performance I/O for PCI- based SCSI, RAID, Fibre Channel, Ethernet, and Gigabit Ethernet adapters, as well as internal RAID controllers. These high performance devices are accessed directly through device drivers in the ESX Server machine and not through a host operating system as with VMware Workstation and GSX Server products.

VMware ESX Server is tested for compatibility with a variety of major guest operating systems running in virtual machines. Additionally, VMware ESX Server is tested for compatibility with currently shipping platforms from the major server makers in pre-release testing. Our goal is to support a variety of storage and network adapters used as standard options for these platforms as they come to market.

VMware ESX Server 3: 802.1Q VLAN Solutions
by VMware on 06/04/2006 @ http://www.vmware.com/pdf/esx3_vlan_wp.pdf

This white paper provides an overview of VLAN concepts and benefits and illustrates three possible ESX Server and virtual machine VLAN configurations. It then compares the advantages

and disadvantages of the three possible configurations and recommends some best practices. The paper also includes configuration samples for both ESX Server and the external physical switches and concludes with a list of frequently asked questions.

VMware Infrastructure 3 architecture
by VMware on 06/05/2006 @ http://www.vmware.com/pdf/vi_architecture_wp.pdf

This paper describes the architecture of VMware Infrastructure, beginning with the elements that make up its physical topology, followed by the virtual, or logical, view of VMware Infrastructure where the relationships between the virtual architectural elements and the physical world are explored. Lastly, the architectures of two core VMware Infrastructure components are discussed in further detail.

Tips and Tricks for Implementing Infrastructure Services on ESX Server
by Foedus on 06/05/2006 @ http://www.vmware.com/pdf/tips_tricks_infrastructure_services.pdf

The purpose of this paper is to provide advice on leveraging key features of VMware Infrastructure for deployment of infrastructure services for system administrators who are new to VMware Virtual Infrastructure. This paper is intended for system administrators who are new to VMware Infrastructure. The recommendations are applicable to small or mid size companies. Recommendations and examples will illustrate what can be accomplished using the VMware Infrastructure platform.

Foedus is a leading provider of Virtualization Infrastructure technologies and services to companies worldwide, specializing in helping organizations utilize emerging technologies to solve their critical business challenges. By offering a unique combination of consulting talent and implementation expertise, Foedus enables companies to realize a measurable return on their VI investment.

VirtualCenter 2: Template Usage and Best Practices
by Foedus on 06/04/2006 @ http://www.vmware.com/pdf/vc_2_templates_usage_best_practices_wp.pdf

This document will focus on the use of virtual machine templates in VMware Infrastructure 3. Virtual machine template functionality was redesigned in VMware Infrastructure 3 which includes VirtualCenter 2. In this context, a virtual machine template (usually referred to as simply a template) is a reusable image created from a virtual machine. The template, as a derivative of the source virtual machine,typically includes virtual hardware components, an installed guest operating system (with any applicable patches) and software application(s). New to VirtualCenter 2 is the ability to keep templates current with OS and application updates.

Foedus is a leading provider of Virtualization Infrastructure technologies and services to companies worldwide, specializing in helping organizations utilize emerging technologies to solve their critical business challenges. By offering a unique combination of consulting talent and implementation expertise, Foedus enables companies to realize a measurable return on their VI investment.

File / Print / DNS Servers: Getting Started with Virtual Infrastructure
by Foedus on 06/05/2006 @ http://www.vmware.com/pdf/file_print_dns_wp.pdf

In this white paper, we will discuss the value of adding virtual infrastructure to the organization to support various staple services such as File and Printing services and DNS. Leveraging VMware ESX Server's iSCSI support, virtual file servers reach new capabilities while keeping costs affordable at the same time. The scalability of virtual networking support in VMware ESX Server 3.0 lends itself especially well in high-bandwidth scenarios when coupled with new processor technology from Intel. New features of ESX Server 3.0 such as clustering, consolidated backup and Distributed Availability Services mean you can ensure 100% availability of file and printing services without the complicated setup of Windows clusters or 3rd party products.

Foedus is a leading provider of Virtualization Infrastructure technologies and services to companies worldwide, specializing in helping organizations utilize emerging technologies to solve their critical business challenges. By offering a unique combination of consulting talent and implementation expertise, Foedus enables companies to realize a measurable return on their VI investment.

Consolidated Backup in VMware Infrastructure 3
by VMware on 06/05/2006 @ http://www.vmware.com/pdf/vi3_consolidated_backup.pdf

VMware Consolidated Backup offloads backup tasks from ESX Server systems to one or more dedicated backup proxies, thus reducing the load on the ESX Server systems and improving manageability. It provides highly flexible backup and restore capabilities, from full image backups (for any guest operating system) to full and incremental file-based backups (for Microsoft Windows guest operating systems).

This white paper provides a brief introduction to VMware Consolidated Backup.

VMware HA: Concepts and Best Practices
by VMware on 01/02/2008 @ http://www.vmware.com/files/pdf/VMwareHA_twp.pdf

This paper describes the use of VMware HA to system and server administrators and IT architects. The paper addresses the important details of the VMware HA architecture and expounds on configuration and implementation topics and how to avoid configuration issues. This paper also discusses various ways that network and storage can be configured and how the choices affect the levels of availability. Troubleshooting and special issues like isolation and maintenance mode are also discussed. Key best practices like basic networking dos and don'ts and advanced VMware HA options and when to use them are explained as well.

Resource Management with VMware DRS
by VMware on 06/05/2006 @ http://www.vmware.com/pdf/vmware_drs_wp.pdf

VMware® Infrastructure 3 provides a set of distributed infrastructure services that make the entire IT environment more serviceable, available and efficient. VMware DRS (Distributed Resource Scheduler) dynamically allocates and balances computing capacity and virtual machine placement with resources pooled from multiple ESX Server hosts.

This white paper provides an architectural and conceptual overview of VMware DRS and describes how you can use DRS to simplify provisioning of applications, achieve higher levels of resource utilization, and better align use of IT resources with business priorities.

Extending Your Patch Management Framework: Technical Note
by VMware on 06/09/2005 @ http://www.vmware.com/pdf/ace_custom_patch.pdf

This technical note explains how to use VMware ACE to extend the patch management

framework that has been adopted by your company.

(script )

Integration with Active Directory
by VMware on 06/09/2005 @ http://www.vmware.com/pdf/ace_active_directory.pdf

This technical note explains how to set up Active Directory to use with VMware ACE.

Custom Authentication: Technical Note
by VMware on 06/09/2005 @ http://www.vmware.com/pdf/ace_simple_authenticate.pdf

This technical note explains how to write a simple script to authenticate remote

VMware ACE users.

(script )

Managing Guest Workers
by VMware on 05/26/2005 @ http://www.vmware.com/pdf/ace_guest_mgmt.pdf

This technical note explains how to use VMware ACE to manage personal computers for guest workers, such as contractors, who routinely use their laptop computers both at work and at home to access the corporate network.

Managing Remote Access
by VMware on 02/13/2005 @ http://www.vmware.com/pdf/ace_remote_mgmt.pdf

This technical note explains how to use VMware ACE to manage remote access through VPN to a corporate network.

Best Practices for Setting up VMware ACE
by VMware on 02/13/2005 @ http://www.vmware.com/pdf/ace_best_practices.pdf

This technical note explains how to set up VMware ACE Manager and most efficiently manage your VMware ACE projects.

Configuring VMware ACE as the Primary Desktop Environment: Technical Note
by VMware on 10/06/2005 @ http://www.vmware.com/pdf/ace_primary_desktop_20051006.pdf

This technical note explains how to configure VMware ACE and a host computer so

the virtual machine running in VMware ACE is the primary desktop environment.

(sample registry file )

Virtual Machine Encryption Basics
by VMware on 01/17/2006 @ http://www.vmware.com/pdf/ace_encrypt_bg.pdf

This technical note provides an introduction to the encryption used in VMware ACE. VMware ACE gives administrators the option of enhancing the security of virtual machines they distribute to end users by encrypting key data and configuration files.

Storage Subsystem Performance in VMware ESX Server: BusLogic Versus LSI Logic
by VMware on 04/04/2006 @ http://www.vmware.com/pdf/ESX2_Storage_Performance.pdf

This white paper compares the performance of virtual BusLogic and virtual LSI Logic SCSI drivers within VMware ESX Server for a variety of data sizes and access patterns.

ESX Server Performance and Resource Management for CPU-Intensive Workloads
by VMware on 02/14/2006 @ http://www.vmware.com/pdf/ESX2_CPU_Performance.pdf

This white paper describes the methods available for allocating and managing resources among ESX Server virtual machines. The paper then examines benchmark results run with various system configurations to illustrate the effects and benefits of using these different scaling and resource allocation methods.

Reference and Capacity Planning with Citrix Presentation Server (for ESX Server 2)
by VMware on 03/09/2005 @ http://www.vmware.com/pdf/esx2_citrix_planning.pdf

This guide recommends architecture designs for users implementing Citrix in a VMware environment. It outlines the advantages of deploying Citrix Metaframe operating in virtual machines running on VMware ESX Server.

Reference & Planning for Virtualizing Citrix
by VMware on 02/22/2005 @ http://www.vmware.com/pdf/esx_citrix.pdf

This provides reference and capacity planning for Citrix Presentation Server in a VMware ESX Server Environment.

Hyper-Threading Support in ESX Server 2.1
by VMware on 04/13/2004 @ http://www.vmware.com/pdf/esx21_hyperthreading.pdf

This paper explains the changes that an administrator can expect to see when running ESX Server on a Hyper-Threading Technology (HT) system.

Best Practices for ESX Server 2
by VMware on 11/04/2004 @ http://www.vmware.com/pdf/esx2_best_practices.pdf

This paper recommends specific deployment guidelines to maximize the ease of use of ESX Server in your environment.

Systems Management
by VMware on 12/06/2004 @ http://www.vmware.com/pdf/ESX2SysMgt.pdf

This technical white paper provides an overview of the systems management options possible with VMware ESX Server 2.

NUMA Support for ESX Server 2
by VMware on 06/10/2005 @ http://www.vmware.com/pdf/esx2_NUMA.pdf

This whitepaper describes how the mainframe-style partitioning of VMware ESX Server 2 and the NUMA system design of the IBM Enterprise X-Architecture complement and support each other.

VMware ESX Server 2: 802.1Q VLAN Solutions
by VMware on 12/10/2004 @ http://www.vmware.com/pdf/esx_vlan.pdf

This white paper provides an overview of VLAN concepts and benefits and illustrates three possible ESX Server and virtual machine VLAN configurations.

Security
by VMware on 09/24/2004 @ http://www.vmware.com/pdf/esx2_security.pdf

This paper describes how VMware provides for security in the ESX Server environment.

CLARiiON Integration with VMware ESX Server
by EMC on 09/14/2006 @ http://www.emc.com/techlib/pdf/H1416_emc_clariion_integration_vmware_wp_ldf.pdf

This technical overview highlights the clustering and VMotion capabilities of the VMware ESX Server when connected to a CLARiiON storage system.

Best Practices Using VMware Virtual SMP
by VMware on 07/12/2005 @ http://www.vmware.com/pdf/vsmp_best_practices.pdf

This paper provides an overview of Virtual SMP and outlines recommendations for deploying virtual machines with more than one virtual CPU on VMware ESX Server.

Timekeeping in VMware Virtual Machines
by VMware on 08/01/2005 @ http://www.vmware.com/pdf/vmware_timekeeping.pdf

This paper describes how timekeeping hardware works in physical machines, how typical guest operating systems use this hardware to keep time, and how VMware products virtualize the hardware.

Backup, Restoration, and Disaster Recovery
by VMware on 11/11/2005 @ http://www.vmware.com/pdf/esx_backup_wp.pdf

Backup, restoration, and disaster recovery are among the most crucial elements of datacenter management. VMware ESX Server System and VMware Virtual Infrastructure provide many different capabilities. This paper describes what resources should be backed up on an ESX server and explains all the options available for that backup.

This paper applies to ESX Server 2.x. The corresponding paper for VMware Infrastructure 3 is available at http://www.vmware.com/vmtn/resources/610

NSI Double-Take and VMware ESX Server & GSX Server Virtual Machines
by VMware on 10/07/2005 @ http://www.vmware.com/pdf/nsi_doubletake_esxgsx.pdf

This document describes how VMware products can be used with Double-Take from NSI Software to provide solutions for challenging high availability and disaster recover cases.

Please see updated paper for VMware Infrastructure 3 at: http://www.vmware.com/vmtn/resources/560

Using IBM® TotalStorage Enterprise Storage Server FlashCopy Function with the VMware ESX 2.5 Server
by VMware on 10/07/2005 @ http://www.vmware.com/pdf/ibm_flashcopy_esx.pdf

In the current information age, the amount of data is increasing exponentially, and storing and managing this vast amount of data is critically important. As part of their data backup strategy, companies are increasingly replicating their data using point-in-time solutions such as FlashCopy® for ESS.

What HA Does & Does Not Do

Fri, 18 Jan 2008 17:58:15 GMT

It is important to understand what VMware HA does and does not provide. What follows is a brief summary of the space.

What HA does

VMware HA covers a growing list of features which are enhanced as new versions are released. New features which often start as experimental become supported for production in subsequent versions. The features of HA are related to a number of different levels within the architecture, as some are related to the virtual machines and some to the ESX hosts.

Note that the real value of HA is that it provides near-zero clustering for ANY operating system and application, with no special licensing (other than the HA license) or special configuration.

Virtual Machine Failure Monitoring

HA features can now be configured for virtual machines in addition to hosts. This is a new feature to VC 2.5 however it is experimental and not supported for production use. By default, Virtual Machine Failure Monitoring is disabled. The feature utilises the HA service so HA must be licensed and turned on for the cluster. VMFM is a global setting for the cluster and settings cannot be changed on a per VM basis.

VMware tools inside the VM sends a heartbeat every second. If the VMFM determines that heartbeats have not been received the VM is restarted.

The elements of VMFM which can be adjusted are

The failure interval at which a machine is declared failed because no heartbeats have been received. Default is 30 seconds.
How long to wait for the machine to start and stabilise before it can be declared failed again. Default is 120 seconds.
The maximum number of failures within a time period before VMFM will stop restart attempts under the assumption that something is wrong with the VM. Default is 3 restarts with no time period defined.

For details on VMFM see http://www.vmware.com/pdf/vi3_35_25_vmha.pdf

Hosts Events

VMware 3i (embeded and installable) monitor the hardware resources of the server. The state of the hardware can be viewed through the Health Status on the configuration tab. Chapter 11 of the 3i setup guide states that the HA service "can take action when hardware events occur". What these events and actions are is currently undefined. However one could reasonably assume that with the increased level of hardware monitoring integration performed by 3i HA might start to take into consideration elements such as CPU or interface card (such as HBA) failure. Lets watch this space.

Host Failure

Host failure is the major function of HA. When the servers in the cluster detect that one or more hosts have disappeared (no heartbeat) it will attempt to restart the virtual machines that were on the lost host in order of priority. Virtual machines may not be restarted because they still have disk locks (meaning host isolation rather than failure) or have been configure to not restart.

Host Isolation

A special case of host failure is host isolation. A host may loose network connectivity to the other hosts but still be running its VMs. In this case the host is isolated causing a "split brain". By default the VMs will be terminated on the isolated hosts, thereby releasing their locks on the disk files. Once the disk files are unlock the VMs will restart on the other hosts. This is often a dramatic event and could cause corruptions in the VM file system. With redundant networking it can be rare of a single host to loose connectivity whilst others have not, therefore the isolation response can be changed to leave the VMs running. In VC 2.5 this can be set at a global level, in earlier versions this must be set on each VM.

What HA does not do

There are a number of elements with the VI3 infrastructure which are not monitored by HA. There can sometimes be a misconception as to the results of certain failures, usually after the event as heads are being scratched wondering what went wrong.

HA does not detect hardware failures within the host, be that CPU, RAM, storage or interface cards such as network or host bus adapters. For this reason it is best practice to include the hardware vendor management agents in the service console and then monitor these with the appropriate platform for your hardware. For example in a HP environment the HP Management Agents for VMware ESX would be installed along with the Systems Management Homepage for Linux . Check with your hardware vendor for their best practices in this area.

Monitoring of the hardware is one of the areas where 3i and its Common Information Model (CIM) architecture introduces both a challenge and an opportunity. The challenge arises in how to monitor the hardware for faults and remote back to standardised management platforms. VMware does state that the CIM is readable by 3rd party management tools (http://www.vmware.com/files/pdf/ESX_Server_3i_presentation.pdf p23). The opportunity is that by VMware working with hardware vendors to integrate hardware state awareness into ESX various services such as HA can utalise this information for increased features sets.

Hardware monitoring is an area where we will see developments over the coming versions.

But I want more

What if you want a greater level of HA, where VM's do not need to be restarted to provide service?

Clusters

Some operating systems support clusters, such as Microsoft Cluster Services (MSCS), which are also supported by VMware (note MSCS is not currently supported in 3.5).

However it's not an uncommon misperception that MSCS maintains application state during a failure. When services are made highly available using MSCS, clients still lose connectivity to the service, the contents of memory is not preserved/copied over to the second cluster node to allow seamless operations. Instead the service is activated on the other cluster node and clients reconnect (in some cases with users not being aware, but not all).

Continuous Availability

VMware demonstrated a future product called "Continuous Availability" at VMworld 2007 that does synchronise application state accross boxes and, given the resources, will allow you to achieve zero downtime in the event of hardware or hypervisor failure.

Provisioning new VM question

Thu, 27 Dec 2007 21:07:54 GMT

A question I have about setting up new VM's in a ESX environment.

Say I have a 3 node ESX cluster. Each box has 2 CPU's and 8gigs of RAM, giving me (6) CPU's and 24gigs total memory.
When people start asking for VM's with requirments such as the need for 2gigs of memory, what starts to happen when you have about 10-15 VM's, and you configure them all with 2gigs of memory? I know they don't use all of it at a given time...but what starts to happen when more and more VM's are built and you start giving 2gigs of memory. Does a future problem arise?

Just wanted to get some feedback on what are some things I should think about when provisioning VM's and maybe what others do.
I sense that the use of resource pools and reservations will play a big part in this.

I really appreciate the help.

ESX does overcommit memory in the sense that you can assign more memory to your vm's in total than it's currently available on the physical hosts. Of course this plays nice when your vm's do not require those 2GB of memory... at least not at the same time. If 20 vm's start to claim 2GB of memory at the same time across 3 systems each with 8GB of memory..... than you <might> have a problem ....

Massimo.

This somewhat depends on your constraints. You can get to a point where you won't be able to start any new vms because your memory is too over committed. A good suggestion is to start out at some point with memory, monitor it and most likely adjust down as you see that memory isn't being used.

Respectfully,

Matthew

Kaizen!

Appreciate it very much.
Anyone have a resource monitor that they recommend? I am also assuming we are talking about measuring resource usage on the VM side of things correct?

Thanks.

TCG

http://www.veeam.com/vmware-esx-monitoring.html

I am not a big fan of spending money unnecessarily, have you check out the performance monitoring inherent in Virtual Center on the host or cluster level yet? These metrics arn't as good as they can be, but arn't horrible either.

Respectfully,

Matthew

Kaizen!

VM does some cool tricks with memory. Yes it will let you over commit, but if you setup say 10 XP machines they can all use the same like memory. Also The nice thing about VMware is after systems run for say 4 hrs. you will see how much memory each system is using. I would not start at 2 GB. I would start at around 1 GB and see if you need to add more. The risk you are going to run by giving them all 2 GB. And this is what you will want to monitor for is called ballooning. The balloon driver tricks the host into thinking it is busy so it can steal some of its unused RAM.

I agree with Milton21. VMware does a good job at reducing the memory footprint across the vm's (although you need to pay attention to the fact that some selected applications tend to use all the memory you assign to them no matter the workload - MS SQL being an example).

I also agree that ballooning is the parameter you want to monitor. You can watch/track it from the VC performance tabs ... as long as it stays at 0 you are fine .... as soon as you see it going up you should either reduce the memory per vm, reduce the vm or add more memory. Look at it as a sort of traffic light .... no by chance the balloon colour code is yellow (or used to be yellow...)

Massimo.

This document was generated from the following thread: Provisioning new VM question

How to decode VML to IBM 2105 ESS-800 LUN ID/serial number for exact order of LUNs

Tue, 04 Dec 2007 18:26:06 GMT

Scenerio;
ESX host 6 only has 1 HBA or HBA #2 is never configured or plugged into the fabric for redundancy during the build. When HBA #2 is configured LUN's may not be presented in the perticular order you desire.
If the exact order is not documented by the storage admin this is how to configure the exact order of the LUN's on the ESX host.

Errors you may find with incorrect LUN order;
LUNs viewed as a snapshot LUN.
Cluster's failing.
Extreme lag time in VIC performance.
HBA's unable to be scanned for volumes/vmfs.
failover connection doesn't work correctly.

Step1; verify volumes by checking UUID
(Volumes do not in every case equal the number of SAN LUNs.)
~~root@vmwesx06~~# esxcfg-vmhbadevs -m
vmhba1:0:4:1 /dev/sdi1 46dfcac8-4a1c7a30-d03f-00145ebd6958
vmhba0:0:0:3 /dev/sda3 470b9baf-d16335c8-cb9f-00145ebd6bb6
vmhba1:0:0:1 /dev/sdb1 46dfca38-305f3580-ee72-00145ebd6958
vmhba1:0:1:1 /dev/sdc1 46dfca81-99901618-7f1b-00145ebd6958
vmhba1:0:2:1 /dev/sdg1 46dfca99-4676f9d8-123a-00145ebd6958
vmhba1:0:3:1 /dev/sdh1 46dfcaaf-432d7528-bcf0-00145ebd6958
vmhba1:0:5:1 /dev/sdj1 46dfcade-854e7510-fbd9-00145ebd6958
vmhba1:0:6:1 /dev/sdk1 46dfcaf2-7fefbb48-ed67-00145ebd6958
vmhba1:0:11:1 /dev/sde1 473de70d-509cd858-c208-00145ebd6958

Step 3;
Check the UUID to match the number of actual volumes to their logical name
(Volumes do not in every case equal the number of SAN LUNs.)
~~root@dol0a1vmwesx06 /~~# ls -l /vmfs/volumes/
lrwxrwxrwx 1 root root 35 Nov 29 06:34 Blade-6-local -> 470b9baf-d16335c8-cb9f-00145ebd6bb6
lrwxrwxrwx 1 root root 35 Nov 29 06:34 Shared-Backup-LUN -> 46dfcaf2-7fefbb48-ed67-00145ebd6958

step 3; verify the SAN lun's you're working with
~~root@vmwesx06~~# more /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: LSILOGIC Model: 1030 IM Rev: 1000
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi1 Channel: 00 Id: 00 Lun: 00
Vendor: IBM Model: 2105800 Rev: .135
Type: Direct-Access ANSI SCSI revision: 03

Storage above is local, where MODEL on SCSI1 shows "2105800" This is the IBM ESS-800 storage LUN.

Now you can't use hparm or sparm to get the information on the drives.
ESS-800 2105 presents LUN's with a XXX-Serialnumber method. In this case it's XXX-27842

step 4;
(This will show you the number of LUNs if you have more LUN's then formatted volumes you can tell here. This is where you can see the difference between the UUID/volumes to LUN from step 1 and 2.)
~~root@vmwesx06 disks~~# cd /vmfs/devices/disks
~~root@vmwesx06 disks~~# ls -l
total 1035553536
-rwx------ 1 root root 73406611456 Nov 29 06:41 vmhba0:0:0:0
lrwx------ 1 root root 42 Nov 29 06:41 vmhba1:0:0:0 -> vml.01000000003231393237383432323130353830
lrwx------ 1 root root 42 Nov 29 06:41 vmhba1:0:1:0 -> vml.01000100003530353237383432323130353830
-rwx------ 1 root root 84000014336 Nov 29 06:41 vml.01000000003231393237383432323130353830

There you see the size and order of the disks. I have a 73GB local drive on vmhba0:0:0:0 and a SAN LUN at vmhba1:0:0:0.
84000014336 Nov 29 06:41 vml.01000000003231393237383432323130353830 is the ESS-800 LUN ID/Serial number.
the number has descriptive digits and 3's appended after each number. Cut it off at the first 3, and the last digit of your serial number.
decode it like this;
vml.01000000003231393237383432323130353830
vml.01000000003231393237383432
vml.0100000000323139327842
vml.010000000021927842
vml.21927842
219-27842

There's your SAN LUN number 219-27842 . Log into your ESS-800 Shark and you can match it up properly.
When your SAN LUN has letters in the serial number the VML number will have a different code in it.
for instance;
vml.01000b00003031433237383432323130353830
vml.0143-27842
01C-27842

Notes;
43 = C
When C in HEX is 12. 4x3 = 12 but it's not hex so that's incorrect.
When you try and divide it the numbers don't add up right. If you find your LUN with an F this is it's code = 0046-27842. When you find a 4 where you would expect a 3 to be located, the number after the 4 determines the letter.
letter = 4 + 1 for each letter to F
A = 41
B = 42
C = 43
F = 46

Now if theres a command to actually match these up as presented on the ESS-800 in the exact order as scanned by the system I have yet to find it. I'm sure someone out there knows an easier method. After you determine your order you can add each LUN to the second adapter's ZONE and rescan after you add each one.

This document was generated from the following thread: How to decode VML to IBM 2105 ESS-800 LUN ID/serial number for exact order of LUNs

Migration from 2.5.4 to 3.0.1 using vmotion with datastore relocation.

Tue, 27 Nov 2007 15:44:20 GMT

Hello,

I am planing to migrate vms to ESX 3.0.1 using vmotion with datastore relocation.

Here I have some questions:

1- How many people have done this ? who had problems? who had not?

2- When migrating with vmotion, the vmdk file on the vmfs2 is deleted after the migration is complete? or it remains on the vmfs2?

3- What if the vmotion fails?

I know the requirement and I also know that I may have problems since 2.5.4 is not supported for this type off "live" migration.

So any comments and help would be appreciated.

Regards

Thanks for your help

I'm not sure that it is possible to VMotion with datastore relocation from 2.54 to 3.0.1.

On these versions of ESX, I believe that you are going to have to perform a cold migration (migrating a powered-off VM) if you want to move the VM to a different datastore. Even then, the different VMFS versions (VMFS2/VMFS3) may cause some issues...Then again, I'm not entirely sure from your description what you are trying to do.

Can the ESX 3.0.1 host "see" the datastore connected to the ESX 2.5.4? If not (and maybe if so) you may have to use WinSCP/FastSCP to move the .vmdk files, then import them with vmkfstools or perhaps use VM Converter.

I think you may want to use a tool called VMWare converter because you won't be able to vmotion.. This will help you migrate your vms better than using the vmkfstools -i format and copy. Also, VMware is now at ESX 3.0.2 Virtual Center 2.0.2. Just an FYI

You also may want to read this site before you begine.

http://www.vmware.com/support/vi3/doc/releasenotes_esx302.html#upgrading

Hope that helped.

1- How many people have done this ? who had problems? who had not?

how many people --> I Dondt know

problems --> some

2- When migrating with vmotion, the vmdk file on the vmfs2 is deleted
after the migration is complete? or it remains on the vmfs2?

You mean DMotion

I dondt use this.

on Vi3

create a new VM, this creates the folder for the vm and than delete the new created VMDK Files (you will deen only this folder and the config files)

when the are able to see both SAN than

use vmkfstools -i /vmfs/sourcevmd.dsk /vmfs/newsan/newvm/newvm.vmdk

with this option you are copying this vmdk when it fails, try again

When the hosts are on different SAN than use fastscp from veeam

attach the vmdk files on your new VM and power on

migrate the VM Hardware, install VM Tools and be happy.

In one of my project i used the original vmx file from ESX 2.x there are some problems. its better when you create a new Vm on Vi3

i Hope this helps you

have a nice day

Nautilius

Well,

we tested it here without any problems, the only things is: Upgrade using Vmotion and datastore relocation is not supported for the ESX 2.5.4 by VMware, that said, it is for 2.5.3.

So we made test on 2.5.4 to 3.0.1 without problems.

We also tested 2.5.4 to 2.5.3 to 3.0.1 since 2.5.3 to 3.0.1 live migration is supported by VMware this could be the way.

But I don't know why 2.5.4 is not on supported list and not on unsupported list on their official website: http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&docType=kc&externalId=8479101&sliceId=2&docTypeID=DT_KB_1_1&dialogID=29898776&stateId=0%200%2029900273

Regards

Jon

1- Ho1- How many people have done this ? who had problems? who had not?

I did this and it worked so well 3 days after the migration was complete I was asked when I was going to start

2- When migrating with vmotion, the vmdk file on the vmfs2 is deleted after the migration is complete? or it remains on the vmfs2?

the files get moved so yes deleted on the VMFS2 partition. Vizioncore has a migrator tool that will leave the VM intact during and after the migration

3- What if the vmotion fails?

then the snapahot is applied and the systems keeps running until you can try again or do a cold migration

Steve Beaver
VMTN Forum Moderator

**Virtualization is a journey, not a project.**

Thanks for your help

Jon

This document was generated from the following thread: Migration from 2.5.4 to 3.0.1 using vmotion with datastore relocation.

Cluster Design

Fri, 09 Nov 2007 13:16:20 GMT

Options with servers regarding HA/DRS

Thu, 18 Oct 2007 14:20:22 GMT

Professionals...

I have a question, in which i could use some real world advice. We have a situation in which our over zealous sales folks have sold a solution promising HA/DRS etc.. to a customer. The customer has a Dell 6850 4x7100 (intel cpus), and a IBM 236 2x intel xeons 3.2 800mhz..

In my mind, those are not the same family, on paper they arent, but can someone confirm my thoughts. I understand that the Xeon 7100's are 800mhz, etc.. but considering that i dont have 2 of those, i cannot comfirm/deny the possibility of those being same family, IE able to support HA/DRS/Vmotion.

Now i can understand the DRS/Vmotion, but with HA, do the procs actually need to be the exact family of processors???

Many thanks for any replies. ALSO, this forum has changed soo much, that i could not navigate it enough to preform a valid search..lol

VMware Communities : Document List - Enterprise Strategy & Planning

Upgrade path from acceleration kits?

VMFS vs RDM (vs VMDirectPath and other solutions)

VM disk type

Performance difference

VMDK disk vs RDM disk

VMDK vs VMDirectPath I/O

VMDK vs native iSCSI

How to build a SAN

Storage Disaster Recovery

SQL Server Clustering Guide

The 2010 Guide to iSCSI Storage

Lab review: Cost-effective iSCSI SANity

How to increase VMware application with shared storage

Disk partition alignment

VI3 on IBM Bladecenter H - High-Level Storage--Datastore Provisioning

VI3 on IBM Bladecenter H - High-Level ESX

VI3 on IBM Bladecenter H - High-Level Blade Network Config

VI3 on IBM Bladecenter H - High-Level Chassis Network Config

ESX on Workstation 6 - Part3 OpenfilerVirtualSharedStorage

Securing VMware ESX servers

Microsoft Cluster Server (MCS) versus VMWare HA

Windows 2000 server bare metal restore from physical server to ESX 3.5 guest OS with Tivoli Storage Manager

VI to VI migration - Howto - Best Practice with minimal downtime

A multi storage vendor post to help our mutual iSCSI customers

Creating task based roles in VMWare

Best Practices for Virtual SAP Infrastructure

Should DC, AD, DNS, DHCP servers be in VMware env?

Small branch office server RAID setup

iSCSI crossover and VI3

IO LOCAL STORAGE ESX All Local Storage environment - performance degradation?

Servie console vmotion network failover and diagram

Teaming NICs in ESX - configuration

Vmotion how to?

capacity bottlenecks

Profiler for VMware: Visibility for Physical and Virtual Worlds

Monitoring Products for VMWare ESX 3.5

Monitoring Solutions for VMware Server Farms

What are my options for Backup of a VM

How do I backup my Virtual Infrastructure?

Some initial statements

Options for backing up Virtual Machines

Running your backup server as a VM

Backup vendors

Specific application suggestions

MS Exchange

MS SQL

Other considerations

References

Which RAID type should I use?

Summary of the different RAID types

Other elements to consider are

Further reading

Document maintenance

List of VMware Whitepapers

General

Server Consolidation

Development and Testing

Business Continuity

Desktop Manageability and Security

Software Lifecycle Automation

VMware Technical Resource documents listing

What HA Does & Does Not Do

What HA does

Virtual Machine Failure Monitoring

Hosts Events

Host Failure

Host Isolation

What HA does not do

But I want more

Clusters

Continuous Availability

Provisioning new VM question

How to decode VML to IBM 2105 ESS-800 LUN ID/serial number for exact order of LUNs

Migration from 2.5.4 to 3.0.1 using vmotion with datastore relocation.

Cluster Design

Options with servers regarding HA/DRS