VMware Communities : All Content - Performance & VMmark

My VMware workstation is very slow!!

pdsouto — Mon, 23 Nov 2009 05:50:41 GMT

Hi Scissor,

First of all thanks for your help.

1. When you asked for reconfiguring my guest to 1 vCPU, you meant to change the number of processors from 2 to 1, isnt it? (sorry for my lack of knowledge).
2. I have changed the memory from 3.5 GB to 1 GB as you said, but the software i am running is quite heavy, it has 200GB. How would be the
maximum amount of RAM I could alocate to my guest? Do you think I would get a better perfmance if I get more memory RAM into this machine?
3. You are right, I am running the software from an external HDD which there is the software pre-installed.

I would like to say your changes were really helful, the performance has increased a lot, it's not perfect yet, but heaps better than before!

Many thanks mate!

Warnings on virtual machine memory usage

walterD — Sun, 22 Nov 2009 10:20:07 GMT

I've installed the latest patch for vSphere 4.
It was also in the answer from MKGuy, but while waiting on reply's, I did not stop searching and found the answer maenwhile.

block level and file level transfer

weinstein5 — Sat, 21 Nov 2009 23:43:41 GMT

No worries - hope it is clear now -

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

Enable trusted execution

aRandomPerson — Sat, 21 Nov 2009 13:09:43 GMT

I have a Dell Optiplex 755, I updated my BIOS settings to Enabled Trusted Execution, VT enabled and virtualization enabled. Now my machine will boot to a blinking cursor. I am unable to get back to the BIOS? Is there a way to resolve this?

Expected vSwitch/vmxnet network performance?

brandontek — Fri, 20 Nov 2009 18:12:11 GMT

Hi, I know this is an old post, but hopefully this will help people get a general idea of vSwitch performance. I just did a test in my lab.

You can view my results here under vSwitch Performance Testing.

http://www.brandontek.com/

How to fetch historical performance of ESx Server using vmware vi sdk

Rajeev S — Fri, 20 Nov 2009 08:30:34 GMT

Hi,

I was trying the script written by ehayes. Script doesnt throw any output, either it doesnt give any output. Guess the .NET datatable is not getting displayed. Below is the output i get,

Collecting: INCH-BGD138 mem.usage.average 05/28/2009 00:00 05/28/2009 23:59
Collecting: INCH-BGD138 cpu.usage.average 05/28/2009 00:00 05/28/2009 23:59

Any help is appreciated.

Thanks

2 vCPUs Hardware Interupts

Wadebum — Thu, 19 Nov 2009 14:03:21 GMT

Kichaonline,

I have put in a support request but unfortunatly our support comes through HP. I don't know if you have anyway of getting them to elevate the SR to you any faster than I can. The HP SR number is 4606821785.

Wade

Performance issue using virtualization

jmattson — Thu, 19 Nov 2009 00:17:11 GMT

After the kudos, it's embarrassing to admit this, but I did all of this testing with Windows 2003 RTM. Windows 2003 SP2 has addressed this particular issue. See this Microsoft TechNet article.

After installing SP2, my new timings are 16 seconds for binary translation and only 6 seconds for VT-x (with or without FlexPriority).

To summarize all of these findings: if you are running this kind of a workload on Windows 2003 pre-SP2, you should use binary translation, but on Windows 2003 SP2, you should use hardware virtualization.

SCSI errors on VMs on NFS datastores

wjs — Tue, 17 Nov 2009 22:53:24 GMT

Any luck with this? I am experiencing the exact same issue. My CentOS images see the problem, but not Windows. My hardware is a bit different however. I assumed in my case, that my NFS server too busy, and I need to move some workloads around.

I need to support an ancient application

oreeh — Fri, 13 Nov 2009 23:52:20 GMT

Depends on the VMware product.
In ESX(i) you can easily limit the guests CPU.

High Ready Times?

mealan — Fri, 13 Nov 2009 21:24:37 GMT

This we thought was very funny when someone was complaining about poor performance, the other 4 4cpu vm's that are all pegged also show a slow down, I hope this makes you giggle.

ESX slow boot at Loading VMKernal IPfc-7.40 (Options")

athlon_crazy — Fri, 13 Nov 2009 08:38:18 GMT

The card in the VMware HCL and latest driver is lpfc_740 version elx_7.4.0.1. BTW, I would start with SAN zoning. Then, from Emulex BIOS, check whether it's able to scan any LUN. If ok, someting may not right with your ESX.

vcbMC-1.0.6 Beta
vcbMC-1.0.7 Lite
http://www.no-x.org

Disk throughput rates on ESX 3.5 and VSphere

newcal — Fri, 13 Nov 2009 08:37:39 GMT

SUBJECT: POOR PERFORMANCE ABOUT I/O SAN DISK WITH ESX 4

If i'm comparing my performance with IO meter with all benches posted in this communities i'm quite good

But i'm agree 200 % with ANDRO... Performance of ESX compared with a physical server are very bad. The 1 Go file transfer test from local ESX to SAN sounds so good to me than using IOmeter because it's real life ...

2.5 minutes for ESX and 15 seconds from a physical server to write a 3 Go file to the SAN: that's a difference

It's not to be admitted, according to me.

Pls find attached my result bench

Regards to everybody

Limit any single process to 25% of cpu

proden20 — Thu, 12 Nov 2009 13:58:57 GMT

I've never used it but check these out:

http://deztec.jp/x/05/faireal/BES-index.html
http://kakku.wordpress.com/2007/10/31/ubuntu-windows-limit-cpu-usage-control-that-bully-ipod-tip/

Dennis Procopio - VCP

Please remember to award points for helpful or correct responses.

Configuration of Hitachi AMS 2100 SAN Storage to VMWare ESX 3.5 with best performance

Seb. Bauer — Wed, 11 Nov 2009 11:21:52 GMT

Hi,

the easiest way to find out why the perfomance is degraded is to have a look in the performance monitor of the AMS2100 which is in Advanced Settings of the Storage Navigator Modular 2.

Then you can analyse where is the problem. High write pending rates are poision for good I/O rates. High usage of single disks or raid groups as well.

Why did you use LUSE LUNs? This is just an advantage when you pick a LUN out of different RAID groups.

Regards

Sebastian

Very slow startup HP 6730b notebook

atom88 — Fri, 06 Nov 2009 14:52:35 GMT

http://www.mcgeown.co.uk/BlogEngine/post/2009/02/18/VMWare-Server-The-VMware-Host-Agent-service-terminated-with-service-specific-error-4294967295-%280xFFFFFFFF%29.aspx

I found this blog which refers to a corrupt datastores.xml file.

However, I do not have this file in the location specified? I'm a little puzzled by this? I also had the error in my windows event logs:

VMWare Server: The VMware Host Agent service terminated with service-specific error 4294967295 (0xFFFFFFFF).

Perhaps this will help others too?

Here's what the post says:

VMWare Server: The VMware Host Agent service terminated with service-specific error 4294967295 (0xFFFFFFFF).

Having just rebooted my laptop, I clicked on the link to open VMWare Infrastructure Web Access in the my browser and was slightly puzzled by the "page cannot be displayed" error. I figured it was probably the VMWare Server Web Access service not starting for some reason, so I opened services.msc and checked it out. The Web Access service was actually running, but the Host Agent service was not, so I tried to start it - it failed. Dutifully opening Event Viewer for some more information I found the error:

The VMware Host Agent service terminated with service-specific error 4294967295 (0xFFFFFFFF).

Great. Thanks for the elaboration. Time to check the VMWare logs, (c:\ProgramData\VMWare\VMWare Server\) - the most recent of which contained the following entry:

~~2009-02-18 10:55:27.908 'App' 4408 info~~ Trying hostsvc
+ ~~2009-02-18 10:55:27.948 'App' 4408 panic~~ error: not well-formed (invalid token)+
+ ~~2009-02-18 10:55:27.948 'App' 4408 panic~~ backtrace:(backtraces not supported)+
+ ~~2009-02-18 10:55:27.948 'App' 4408 info~~ Win32 service stopped+

A bit of Googling pointed me in the right direction, "not well-formed" refers to the XML config files which are stored in c:\ProgramData\VMWare\VMWare Server\hostd, after that it was just a case of opening each one in turn until I found the malformed one. I backed it up and replaced it with a default version, and lo and behold, my services started!

fantastic stuff.. this plagued me as well.. turned out to be the datastores.xml fine was corrupted/empty. There was however another version of the file called datastores.xml.default. so i just renamed that to be datastores.xml and started the VMware Host Agent. Perfect, Thanks!!

best

Monday 30 March 2009 16:02

had the same problem, corrupt datastores.xml file, moved it out and started the VMware Host Agent.
Worked a charm....Thanks!

mack

Thursday 11 June 2009 10:16

Thanks to all of you for taking the time out to share this with all of us! This fixed my problem within minutes!

VM object in perfmon not listed?

scott28tt — Thu, 05 Nov 2009 13:46:06 GMT

Close and re-open Perfmon (the VM objects sometimes appear, sometimes don't), or try re-installing VMware Tools.

Scott.

Blog: http://vmwaretraining.blogspot.com
Web: http://www.thinkvirtually.co.uk
Twitter: http://twitter.com/vmtraining and http://twitter.com/thinkvirtually

Virtual Desktop Performance issues - ESX 3.5 VM View

ciberweb — Wed, 04 Nov 2009 17:26:43 GMT

Sorry, wasn't sure what category to submit this under.

I have a small environment running 1 VM server and 5 VM Desktops. Performance on the Desktop side is the issue.

Server Hardware:
1 - Dell 2950 / 2-Quad 2.8ghz Proc / 16gb RAM / 1.2TB of local storage - 15k RPM HD's / 2 - Dual NIC's (1 NIC = 2 VM Desktops, 1 NIC = 3 VM Desktops and 2 NIC = Server)

Software:
Server -> SBS 2008
PC's -> Windows XP Pro SP3 - Office 2003, ERP Application, File Maker, Adobe and WinZip

Virtual Hardware:
1 2.8ghz Proc / 1gb RAM / 25gb C:\ drive (15gb free)

Physical Hardware used for Virtual PC's:
Dell Dimension 2400 -> 2.4ghz Single Proc, 512mb or 768mb RAM, 100mb NIC

Issues:
VM users are seeing slowness issues when attaching documents to emails in Outlook.
Refresh is slow when changing from one email to another.
Working in one application and then toggling to another hangs the desktop for a bit.
Generally speaking applications seem to be slow.

What is my bottleneck? Is this typical or is there a config change that can be made to resolve the issues?

Help would be greatly appreciated.

Marc

Advanced Networking Performance Options

drummonds — Wed, 14 Oct 2009 16:58:33 GMT

Some of the advanced networking options available in vSphere 4.0 are reviewed in this paper. Many of these options control trade-offs between latency, throughput, CPU utilization, and reliability (e.g., dropped packets). It is not possible to optimize all of these at the same time, so option defaults are chosen to be suitable for the vast majority of applications. These options are provided to meet the stricter requirements of other applications. Advanced options often have subtle side effects, or merely move an issue from one area to another. Therefore it is recommended that VMware Support be engaged before changing such options, especially for production machines.

There are over 100 options that can be set under Configuration → Advanced Settings → Net. Of these, the ones listed below are most likely to be useful for tuning networking performance. Many of the others are for internal testing or enable unreliable features.

All of the options listed here take integer values. For the “Boolean” ones only the default value is shown: 0 for “false”, and 1 for “true”. Other parameters are shown with their default, minimum, and maximum values.

Parameter Name	(Default, Minimum, Maximum)	Description
MaxPortRxQueueLen	(80, 1, 500)	Maximum length of the Rx queue for virtual ports whose clients support queueing. Possibly should be increased if Rx packet drops are seen in the port connected to a VM. Relevant only for e1000 vNICs used with Fault Tolerance (FT) and VLANs.
MaxNetifTxQueueLen	(500, 1, 1000)	Maximum length of the Tx queue for the physical NICs. Increase if Tx packet drops are seen in uplink port to the pNIC.
GuestTxCopyBreak	(64, 60, 4294967295)	Packet header transmits smaller than this in bytes will be copied rather than mapped. More security and functionality than performance implications.
VmxnetTxCopySize	(256, 0, 4294967295)	Transmits smaller than this in bytes will be copied rather than mapped. Copying costs CPU but puts lets pressure on the Tx queue and doesn’t require completion.
VmxnetWinUDPTxFullCopy	(1)	Enable full copy of Windows vmxnet UDP Tx packets. Might disable to save CPU, especially for jumbo frames, at the cost of risking more packet drops.
NetTxDontClusterSize	(0, 0, 8192)	Tx packet size (in bytes) smaller than this are transmitted immediately (coalescing options are over-ruled for these packets). Used to ensure good latency for small packets.
CoalesceTxTimeout	(4000, 1, 4294967295)	The coalesce timeout in micro-seconds, or effectively the maximum latency without transmitting. Smaller values can reduce the packet latency at the cost of CPU. Risky to go below 1000.
CoalesceDefaultOn	(1)	Enable dynamic coalescing. Disable to test if issues are related to coalescing.
CoalesceHandlerPcpu	(1, 0, 128)	pCPU that coalesce timeout handler runs on. May be important to set this if VM CPU pinning is used.
CoalesceTxQDepthCap	(40, 0, 80)	Maximum number of “normalized” Tx packets to coalesce. Reduce if Tx coalescing appears to be too aggressive.
CoalesceRxQDepthCap	(40, 0, 80)	Maximum number of “normalized” Rx packets to coalesce. Reduce if Rx coalescing appears to be too aggressive.
vmxnetThroughputWeight	(0, 0, 255)	How far to favor Tx throughput for vmxnet 2 & 3. “0” is dynamic, otherwise this is a weight where a lower value favors latency and a higher value favors throughput.
TcpipHeapSize	(24, 24, 120)	Initial size of the TCP/IP module heap in megabytes. May need to increase if there are many vmkernel connections (NFS, iSCSI, etc.).
TcpipDefLROMaxLength	(16000, 1, 65535)	Maximum length for the LRO aggregated packet for vmkernel connections. Increasing this reduces the number of acknowledgments, which improves efficiency but may increase latency.
E1000TxZeroCopy	(0)	If disabled copy UDP or non-TSO Tx packets for e1000.
E1000TxTsoZeroCopy	(1)	If enabled do not copy TSO Tx packets for e1000.
E1000IntrCoalesce	(1)	Enable interrupt coalescing for e1000. Disabling can improve latency at the expense of CPU.
MaxPktRxListQueue	(3500, 0, 200000)	Maximum number of packets queued in vmkernel. Increasing this can reduce the number of dropped packets but at the cost of increased vmkernel memory and queuing latency.
Vmxnet3RSSHashCache	(1)	Enable RSS hash cache for vmxnet3 in Windows guests.
VmklnxLROEnabled	(0)	Enable large packets for recent Linux guests with vmxnet 2 & 3. Most likely to benefit hosts with small number of VMs with few sessions each, where each session has a heavy Rx load (more than 1 MB/sec). This is an experimental feature and has not been tested extensively.
VmklnxLROMaxAggr	(6, 0, 24)	Maximum aggregation count in number of packets for vmklinux LRO.

creating vm pc's?

KNardi — Tue, 03 Nov 2009 16:28:37 GMT

VSphere 4.0, VCenter 4.0, ESX hosts = 9 all using HA and DRS on one Cluster

My question is, Is it possible to deploy a PC workstation using LANdesk on the VIC?

Can we use a PXE boot?

IOMeter testing within the Service Console

oschistad — Tue, 03 Nov 2009 11:30:27 GMT

As a previous poster also mentioned, running IOmeter or any other sort of benchmark from inside of the service console will not give you any useful data. Firstly, the SC is actually a virtual machine and has to go through the vmkernel to perform disk IO just like any other VM, and secondly the SC is intentionally limited in how much I/O it is allowed to generate. So if you were to run IOmeter you would see that your benchmark runs from inside the SC was much slower than what you'd see from a regular guest VM.

Disk performance strangeness (Guest OS slow, ESX storage blazing fast)

oschistad — Tue, 03 Nov 2009 09:00:47 GMT

Well, in our case it's actually the buslogic driver which gives us good performance while the lsilogic parallell scsi runs like molasses.

AMD Opteron 2435 2.60 GHz Six Core VS Quad Core AMD Opteron(tm) 8384; 2.7GHz

fanzalone — Mon, 02 Nov 2009 21:31:57 GMT

Dear All,

I'm new in the vmware community.

I'm using vmware Enterprise 3.5 since 3 years.

I need to change two of my vmware hosts and I would like to know which benchmark score this two different machines have.

both machine are dual socket equiped.

Thanks a lot for your help.

Memory Reporting: who do I trust - vCenter or Windows PerfMon?

Dave.Mishchenko — Sat, 31 Oct 2009 10:05:00 GMT

The guest will have better visibility to how active it's memory usage is. Guest active memory is an approximation by ESX(i) of what it thinks the guest is activitely using for memory. It can take time to catch up to the guest OS and shouldn't be a concern unless the number begins to approach the memory assigned to the guest. As an application requests and then frees memory, the guest OS will track this accurately, but the management of allocated and freed memory is not passed to the hypervisor level.

In regards to the guests using 100% memory, it believes that it has physical memory and thus will allocate memory as it sees fit. Should it approach it's limit it will begin to swap to disk just as guest OS would on a physical host. What is different from a physical host is that ESX(i) will use transparent page sharing, ballooning and if necessary swapping to provide memory to all the guests running on it.

Dave
VMware Communities User Moderator

New book in town - vSphere Quick Start Guide -http://www.yellow-bricks.com/2009/08/12/new-book-in-town-vsphere-quick-start-guide/.
Do you have a system or PCI card working with VMDirectPath? Submit your specs to the Unofficial VMDirectPath HCL - http://www.vm-help.com/forum/viewforum.php?f=21.

What VMware product to use? Your dream setup high-powered personal workstation...

EricBryant — Fri, 30 Oct 2009 20:20:47 GMT

Hello All,

If you want to run multiple guest O/S's for your personal workstation, and want the absolute highest possible desktop performance ever (including graphics performance) and you had a big beefy host with plenty of CPU & RAM to put it on, would you run vSphere, ESXi, or Workstation 7 on it (all cost considerations aside) and if you chose Workstation, what base O/S would you use assuming you want maximum RAM available to guests?

Thanks!

Eric

VM suddenly running VERY slow

mattstewartcsd — Fri, 30 Oct 2009 14:34:41 GMT

They were identical to what they are now. It's very odd. It was almost as if the server itself was ok, it was just the Windows GUI that taking was taking enternity to respond.

Either way, the move has obviously kicked it back into life. Thanks for your help.

vSphere vcpu scheduling imbalance

ltbraswell — Thu, 29 Oct 2009 19:16:11 GMT

My group has been doing some testing comparing virtual to bare-metal performance with a particularly high-compute utilizing application.

Platform: Intel E5540 (Nehalem) quad-core, dual-socket. 48GB of 1066MHz memory (properly balanced)

OS/guest OS: RHEL 4.7 (for both bare-metal and virtual configs)
Configuration: Hyper-threading off for both esx and bare-metal. Numa enabled and verified on both ESX and RHEL 4.7 (bare-metal)
ESX version: 4.0u0 build 164009
ESX host power.CpuPolicy is set to "static"

The application is a single-threaded modeling/simulation type program. It is not I/O intensive at all and basically each job generates a load average contribution of 1 (i.e. it pegs a cpu for each job). What we have been doing is running 1,2,4, and 8 simultaneous jobs and comparing the virtual to bare-metal physical performance with different combinations of VM vcpu counts (1,2,4 vcpu). Don't know a lot of details about the application other than when scaling from 1 to 8 jobs on bare-metal, the performance of any one job degrades somewhat so it seems to be memory constrained (not sure if the is sensitive to last level cache size or raw RAM performance). The "model" size itself for each job is 1.5GB.

The observation is that when simultaneously running 2 and 4 of these applications, each in its own 2vcpu virtual (so 2 and 4 virtuals respectively but all running on one 2-socket physical). More often that not for the 2 job case, we will see the worlds representing these active vcpus both running on 1 socket (with the other socket basically idle). For the 4 job case (though not as frequently), we have seen the load split 3/1 across the 2 sockets. When we run 8 jobs we get an even distribution of active vcpu worlds across the 2 sockets (4 on each).

The question is: Why is it that when there are 2 active vcpus does ESX schedule these vcpus on one socket more often than not. Same thing for 4 active vcpus, lopsided scheduling 3/1, though not as frequently. The hypothesis is that this behavior is related to some power management function of the scheduler (i.e. it would rather load up one socket more fully so that the cores on the other socket have a chance to be freq scaled down or in the halted state and therefore save power).

Ordinarily for an application that is not sensitive to memory performance or last level cache size this might not be an issue but for our application it apparently is sensitive the load being spread across sockets.

We can workaround this lopsided scheduling by using affinity rules to pin virtuals to sockets and manually force the load to be spread but we were hoping for some way to get it to balance the load without using affinity rules. Note that our hosts are running with power.CpuPolicy static.

Any ideas?

Message was edited by: ltbraswell changed subject to remove the reference to version 4

ESXi 4.0 shows incorrect memory usage at host level

dasbacgl — Thu, 29 Oct 2009 17:33:56 GMT

Hey,

I've installed the Nehlem patch on my ESXi 4.0 boxes (current build #193498). The individual guest VM's now report their Active memory correctly, and Memory Usage is also correct, however at the host level it now appears that ESX is reporting Memory Usage (%) based on Consumed memory instead of Active Memory.

Has anyone else encountered this?

Thanks,

Garrick

When to add additional memory to VM?

TobiasKracht — Thu, 29 Oct 2009 11:31:38 GMT

I mean that oveall amount of your RAM is 4 Gb and you config VM to use 2 Gb, your system may run slow.

StarWind Software R&D
http://www.starwindsoftware.com

Performance

nellore — Thu, 29 Oct 2009 05:02:58 GMT

HI,

I installed ESX3.5i and created 2VM's , booting of the VM's are very slow

1. HP blade 8.GBRam, 2.6CPU assigned 3.5GB for VM still it is taking 16min's to boot to window's
configured with windows2003 with SP2,Etrust antivirus,with updates and windows latest Update

2. When ever we take base Server Restart NTP service is getting Down.

3. I am facing in VM's Network problem using Netbios copy it is taking around 14mins has taken 64mb file

Kindly let me know Anyone has faced this issue.

Anyone familiar with STPNavigator tool

RDellimmagine — Thu, 29 Oct 2009 01:03:40 GMT

Sorry, I'm not familiar with this tool at all. But this is the community (or possibly the vSphere storage community) where knowledgeable people are.

Regards, Robert

Windows 2008R2 Terminal server - very bad performance

aandriolli — Wed, 28 Oct 2009 21:19:54 GMT

Windows 2008 R2 is still not supported on ESX 4.

VMs Made in Brazil
PS: por favor considere dar pontos a este ou qualquer outro post caso lhe seja útil.

Keyboard and Screens slow to respond

aandriolli — Wed, 28 Oct 2009 21:00:02 GMT

Try following recommendations at

http://virtrix.blogspot.com/2007/03/vmware-best-practices-for-deploying.html
http://viops.vmware.com/home/docs/DOC-1226

Also, for Terminal Services VMs you might consider installing the VM from zero instead of P2V:

http://communities.vmware.com/thread/193845

VMs Made in Brazil
PS: por favor considere dar pontos a este ou qualquer outro post caso lhe seja útil.

Capacity Question

vmroyale — Wed, 28 Oct 2009 12:12:12 GMT

Hello.

Can you describe more about the hardware and software you plan to use? VMware Server on a 3 disk RAID 5 array will have very different scalability in comparison to ESX with shared storage. If you just need general ideas, check out the VMware Calculator or the HP VMware Solution Sizer.

Good Luck!

Monitoring Memory Usage

TobiasKracht — Tue, 27 Oct 2009 14:49:47 GMT

You run P2V for Windows? Did you clean already unpresent devices after P2V? VMTools installed? Also look at this http://communities.vmware.com/docs/DOC-5430

StarWind Software R&D
http://www.starwindsoftware.com

opinions on the 10gb network cards

cxo — Mon, 26 Oct 2009 19:19:49 GMT

Would like to know people's opinions on the 10gb network cards. In particular the "Intel 10 Gigabit AF DA Dual Port Server Adapter" and the "Chelsio Communications S320E: 10 Gbe Nework Adapter".
Just looking for horror stories.

San Sizing documentation request...

jjkrueger — Mon, 26 Oct 2009 12:34:16 GMT

That's the trick with disk sizing - there's no such thing as a boilerplate recommendation.

Your LUN sizing will most often be dependent on the I/O characteristics you need for the device - IOps, disk throughput, etc. The number of VMs you can reasonably place on a VM will also be limited by those numbers. What I try to do when sizing my environments is figure out a baseline of peak (think worst-case application scenario here - what if all of the VMs' apps will be going nuts on the disk at the exact same time) IOps (that's usually the biggest disk bottleneck - physics are a pain here, as the head on a spindle can only be in one place at a time), figure out how many spindles I need to meet that. Once I know the number of spindles my LUN has to be spread across, I can look at the overall size of the datastore that I'll need. Keep in mind not only disk sizing here, but also whether or not VM snapshots will be put to use and for how long, and also keep in mind the memory being allocated to each of the VMs to plan for .vswp file sizing. Once I have sizing information, I can look at the device split size on my storage array, figure out how many splits I need to meet both the I/O and size requirements for the data store, and then create the LUN keeping in mind RAID overhead. A rule of thumb I still use is 1 physical spindle provides me with ~150 IOps to work with. That number is very conservative (and frankly a bit old, but I haven't looked at disk in too much depth in the past 3 years). Also keep in mind that your LUNs will be sharing spindles (and therefore, sharing IOps) with other LUNs, so other activity on the storage array needs to be taken into account during the design process as well.

For example, If I have an environment with 10 applications, each with a peak IOps of 500, I would ultimately want a LUN that would support 5000 IOps. Working with my 150 IOps rule of thumb, I would be looking at a LUN spanning 34 spindles. If each of the VMs was provisioned with 50GB virtual disks and 4GB of RAM each, and I knew the VMs would not have any point that the snapshots would grow beyond 10% of the original virtual disk size, I would plan for 550GB for virtual disk/snapshot space, 40GB for .vswp files, and some overhead for VMFS metadata, config and log files, etc. So I'd be looking at, say, a 600GB LUN. At this point, can my storage array support a 600GB LUN spanning 34 spindles? That's a 17.6GB split size. Some compromise may need to be made depending on what you know of your application performance (is peak load going to happen all at once, or is the worst aggregate peak load lower? that could lead to a larger split size and fewer spindles) or what you know of your storage array (split sizes are available larger and smaller than what the calculations show - do you go with a smaller split and more spindles, wasting available IO, or do you take the larger split size with fewer spindles, with the possibility of either starving yourself of IOps or wasting precious disk space).

As for maximum numbers of VMs placed on a given data store, you wouldn't want many more than 32 VMs on each, largely due to the way VMFS file locking is maintained. Each host will lock files in the metadata space for the VMFS datastore. The locks are updated periodically. Those updates, and any other metadata update activities, cause an entire SCSI device lock to be maintained on the disk device. Above 32 heavily used VMs, metadata updates can easily keep the LUN in a locked state, preventing VM disk I/O from occurring.

It's nearly impossible to put together a sweeping document because recommendations made for your environment are specific to your environment. Sure, workloads between dissimilar shops are similar, but the subtleties of each shop means that any generalized recommendation like "build 500GB LUNs across 10 spindles and put 15 VMs on them" would be overkill for 45% of the shops looking, and would be a performance disaster for 45%, leaving a smallish 10% of environments happy and in a good comfort zone.

Hope that makes some sense, and helps.
-jk

vmware server performance benchmarks...

eggie — Mon, 26 Oct 2009 02:41:11 GMT

Thanks for the response. Are there any other metrics to follow other than vmmark scores?

Resource Pools

Mr G Grant — Fri, 23 Oct 2009 11:24:43 GMT

They each have 64GB.

The average is about 1.5 GB per VM. Some machines have as much as 8GB and i'm sure this has been over provisioned.

Does anybody have a good working practice of how to monitor over provisioned virtual machines? I would love to be able to claim some memory back.

A Newbie needs help in a big way....HELP!!!

Dinube — Thu, 22 Oct 2009 21:19:50 GMT

Aloha:

I just installed VMware-Fusion-2.0.6-196839.dmg on My MacBook Pro running OS 10.5.8. It seems to work well.

However I am having problems installing a Hanger 9 simulator. It's on 4 separate disks. The first disk goes on ok asked for disk 2, but does not eject disk one.

I get an error panel which is attached.

Any ideas??

Thanks
Dan Page

VA deployment degradation on vSphere4

Lexus16 — Thu, 22 Oct 2009 11:08:21 GMT

Hello,

We have high performance degradation when our Appliance is deployed to ESX server via Virtual Center. When the appliance is deployed to ESX directly we don’t have such problem.
We are using the following script:

curl -u username:NfelMC2Ub -k -T upload https://10.250.148.27/folder/LR1_11133_ESXAppliance/upload.dat?dcPath=Center&dsName=Storage1

where 10.250.148.27 is the address of Virtual Center. If we specify the address of ESX instead of Virtual Center, everything works fast.
BTW such problem appeared on vSphere 4, on previous versions everything was fine.
Could you please answer:

What could be the reason of this degradation?
How can we tune up the speed of deployment via Virtual Center?
If we have correct credentials on Virtual Center, is it possible to deploy directly to the ESX registered on it?

Thank you

IRQ interrrupts

NTurnbull — Thu, 22 Oct 2009 09:34:15 GMT

Yep, the IRQ is the number in the first column, so your host usb controller(s) are using IRQ 20 and 21 which isn't showing as sharing as per your first screenshot.

Thanks,
Neil

Recommanded VM instances in 1 esx host 3.0 and 3.5

sean051 — Thu, 22 Oct 2009 03:13:17 GMT

I feel for quality performance...max 20 vm instances per esx host..of course its also depends on how powerful the esx host..

I wonder if there are any calulation for these ratio btw esx host and VM instances.

Using vscsiStats for Storage Performance Analysis

drummonds — Fri, 29 May 2009 22:28:27 GMT

Introduction

esxtop is a great tool for performance analysis of all types. However, with only latency and throughput statistics, esxtop will not provide the full picture of the storage profile. Furthermore, esxtop only provides latency numbers for Fibre Channel and iSCSI storage. Latency analysis of NFS traffic is not possible with esxtop.

Since ESX 3.5, VMware has provided a tool specifically for profiling storage: vscsiStats. vscsiStats collects and reports counters on storage activity. Its data is collected at the virtual SCSI device level in the kernel. This means that results are reported per VMDK (or RDM) irrespective of the underlying storage protocol. The following data are reported in histogram form:

IO size
Seek distance
Outstanding IOs
Latency (in microseconds)
More!

Running vscsiStats

vscsiStats collection and analysis requires two steps:

Start statistics collection.
View accrued statistics.

Documentation on command-line parameters are available when running '/usr/lib/vmware/bin/vscsiStats -h'.

Starting and Stopping vscsiStats Collection

The tool is started with the following command:

/usr/lib/vmware/bin/vscsiStats -s -w <world_group_id>

This command starts the process that will accrue statistics. The world group ID must be set to a running virtual machine. The running VMs' IDs can be obtained by running '/usr/lib/vmware/bin/vscsiStats -l'.

After about 30 minutes vscsiStats will stop running. If the analysis is needed for a longer period, the start command should be repeated above in this window. That will defer the timeout and termination by another 30 minutes.

Since results are accrued and reported out in summary, the histograms will include data since collection was started. To reset all counters to zero, run '/usr/lib/vmware/bin/vscsiStats -r'.

Viewing Statistics

Counters are displayed by using the following command:

/usr/lib/vmware/bin/vscsiStats -p <histo_type> [-c]

The histogram type is used to specify either all of the statistics or one group of them. Options include all, ioLength, seekDistance, outstandingIOs, latency, interarrival.

Results can be produced in a more compact comma-delimited list by adding the optional "-c" above.

Using vscsiStats Results

Use Case 1: Identifying Sequential IO

Storage arrays can process sequential IO much faster than random IO. You can therefore improve the performance of a sequential workload by placing it on a dedicated LUN to allow the array to optimize access. vscsiStats can help you identify your sequential workloads even if you don't understand anything about the application in the VM.

Take the following graph as example, which I generated by running '/usr/lib/vmware/bin/vscsiStats -p seekDistance':

This graph shows that most of the commands are being issued a great distance from the previous command. It looks like all of the commands were 50,000 or more logical blocks away from the previous command. When I looked at the raw data, I saw that over 99% of the commands were more than 128 blocks away from the previous command. That's random access if I've ever seen it. Here's the opposite example:

In this case the logical block number (LBN) of each command is most frequently exactly one larger than the previous command. That's the signature of a heavily sequential workload. It shouldn't surprise you to learn that both of these profiles were generated by Iometer using random and sequential writes, respectively.

Use Case 2: Optimizing for IO Sizes

The IO size is an important characteristic of storage profiles. A variety of best practices have been provided by storage vendors to enable customers to tune their storage to a particular IO size. As an example, it may make sense to optimize an array's stripe size to its average IO size. vscsiStats can provide a histogram of IO sizes to help this process. The following graph was generated by '/usr/lib/vmware/bin/vscsiStats -p ioLength':

From these results I can see that about a quarter of the commands came in IOs smaller than 4k. About half of the commands were sized to 4k commands. The minute number of remaining IOs were larger than 4k. This signature is common of a VMDK formatted to 4k blocks and supporting OS and application execution. The storage array should be optimized for 4k blocks if this disk's performance is a priority.

Use Case 3: Storage Latency Analysis (Including NFS!)

esxtop is a terrific tool for latency-based storage analysis. Fibre Channel and iSCSI HBAs have device and kernel latencies in esxtop's storage panel. Software iSCSI initiators will show up as vmhba32 (ESX 3.5 and earlier) and vmhba33 (ESX 4.0 and later.) But esxtop does not provide latency statistics for NFS stores.

Because vscsiStats collects its results where the guest interacts with the hypervisor, it is unaware of the storage implementation. Latency statistics can be collected for all storage configurations with this tool.

The above graph shows that the server in my office with a single direct-attached SCSI disk is performing as I would expect. About half of all the operations are completing in under 5 ms. The other half take 5-15 ms to complete. A few commands took longer than 15 ms, but the number is so small that it doesn't concern me. Similar results can be seen with NFS arrays.

vscsiStats on ESXi

vscsiStats can be installed on ESXi hosts after putting the host into tech support mode. More information on this process is availalble on Scott's blog on the subject on vPivot.

Additional Resources

My colleagues Ajay Gulati, Chethan Kumar, and Irfan Ahmad presented at VPACT 09 Storage Workload Characterization and Consolidation in Virtualized Enviornments. This paper serves as an excellent example of vscsiStats in action.

I learned vscsiStats by reviewing Irfan's VMworld 2007 presentation (vscsiStats: Fast and Easy Disk Workload Characterization on VMware ESX Server) and playing with the tool. Check out his presentation if you'd like more detail.

Listing hosts from a failed VM ESX 3.5 server crash

Just_Bob — Tue, 20 Oct 2009 21:41:20 GMT

In our instance, all of the servers were migrated to other ESX servers and the one that crashed was put in maintenance mode for diagnosis. We have about 15 ESX servers with about 200 hosts. We need to know what servers were on the ESX at a point in time.

Slow Network for 1000/Full Host, Fast Network for Guests?

Jamie Rix — Tue, 20 Oct 2009 19:42:48 GMT

I know this is an old thread, but the descriptions here are indicating that it might be a shared IRQ issue that is affecting the throughput on the service console nic.

KB1003710 has all the info.

Degradation when snapshots are deleted?

esiebert7625 — Tue, 20 Oct 2009 18:23:08 GMT

No problem, snapshots are often misunderstood, especially by those new to VMware. Once you start playing around with them and getting to know more about them it becomes easier to understand how they work and the impacts of using them.

Eric Siebert
VMware Communities User Moderator
-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-
Author of the upcoming book "VMware VI3 Implementation and Administration" now available on Rough Cuts
Check out my website: VMware-land
Read my virtualization blog: SSV Blog
-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Performance issue when deleting snapshot

chrisAMS — Tue, 20 Oct 2009 14:51:24 GMT

Thanks, i have already this configuration.

ESX 3.5 VM Perfomance Charting

vmproteau — Mon, 19 Oct 2009 16:53:45 GMT

ESX 3.5 (build 163429) Virtual Center 2.5 (build 147633)

In our environment, we have a predominance of WIndows servers. When asked for VM performance data, we typically recommend the client perfom OS level performance monitoring for detailed information. There are an increasing amount of requests where Virtual Center metrics and performance charts suffice and I see us getting more requests to provide this level of perfomance data and charting.

When charting in Virtual Center, I notice that realtime data has more metric options than day, week, month, year and custom. Particularly in disk, outside of real-time, I only have a disk usage metric.

First, is what I descrie what I should expect to see?
Is there a configuration option (or detail option) I can set to allow those metrics to be chartable in the other time frames?
If there is, is it wise to make that change (will it have potential negative effects)?

Any information about performance charting for VMs withing Virtual Center would be helpful.

Microsoft Dynamics NAV/Windows Server 2008 performance problems

TobiasKracht — Mon, 19 Oct 2009 11:38:40 GMT

Except setting JumboFrame to 9k you also have to change following registry parametrs in HKLM\System\CurrentControlSet\Services\Tcpip\Parametres:

GlobalMaxTcpWinowSize=0x01400000
TcpWindowSize=0x01400000
Tcp1323Opts=3
SackOpts=1.

All paramets are DWORD type, and after applying you have to reboot server.

StarWind Software R&D

ESX SAN Sizing (HA, VMotion)

AndreTheGiant — Mon, 19 Oct 2009 05:03:03 GMT

The C:/D: thery could be used but not necessary will improve your I/O speed.
You must check on your storage documentation if there are some best practices for RAID, LUN, multipath and so on...
You can split OS / Data disk only for guest that really need this.

Andre

Centos 5 VM 100% CPU usage

godpickny — Mon, 19 Oct 2009 05:02:08 GMT

Hi, sorry for the late reply, been busy preparing for my wedding.

I did what you asked...installed vmware tools and messed with the time settings so now I'm in a wait and see mode

By mid week I should know if the changes made any difference. The server sofware is Cumis, its a database for Credit Unions

from a Trinidadian company named MSD, Micro Software Design.

Thanks much for your help...I'm trying to get a hold on the features of ESXi, seems a very useful tool.

reinstalling after time machine back up

phydaux — Sat, 17 Oct 2009 10:16:56 GMT

Hi there,

I'm about to replace the hard drive in my mac book, I plan to reinstall everything
from time machine.

Has anybody done this? are there any special issues with VMware or XP, will all my XP
data be OK or should I reinstall it separately?

Thanks in advance.

PS I’m running Mac 10.5.8 VM ware
2.00 Windows XP.

Is ESX a lot better than vmware server in performance??

drummonds — Fri, 16 Oct 2009 20:54:15 GMT

Yes. ESX (as well as ESXi) is much better than VMware Server. Check out this graph from some old data the perf team generated for VMworld a couple of years ago:

This is basically equivalent systems running VMmark with either ESX or Server. I believe these results were done on ESX 3.0.1. ESX performance has improved much more than Server in the past two years, too.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

IBM LS22 server and Windows 2008 Terminal Server CPU Maxing Issue

Nemus — Fri, 16 Oct 2009 16:31:33 GMT

I am running VM4 Sphere with a guest OS of Windows 2008 Terminal server on an IBM LS22 with 16 gigs of ram and 2 cpu's after the fith person logs in to the Win 2008 Terminal Server the cpu hits 100% on the virtual machine.

My assuption is that its an issue with Windows 2008 TS ,but thought I would do a post here to see if anyone could shed some light on the issue.

The system runs to a crawl after this happens and I haven't been able to run any other virutal machines.

This server is Runing VM V4 sphere

I've been runing V 3 with out an issue, but the new Sphere V4 seems to be buggy.

Any suggestions about what could be causing this issue would be helpful.

So to recap my questions are:

Is this a vmware issue or just an Guest OS issue?

If it is a vmware issue what could be causing the guest OS to spike to 100% on both cpus?

Thank You

Memory balooning & performance

drummonds — Thu, 15 Oct 2009 22:45:31 GMT

We recently released a paper on ESX memory management, which included balloon and swap performance. See my blog on the subject.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

IOmega ix4-200d IOmeter results (100MB/Full Network)

khughes — Thu, 15 Oct 2009 21:07:59 GMT

10G Network Performance - max 1.6Gb/s

jpdicicco — Thu, 15 Oct 2009 15:11:35 GMT

Have you checked your server to verify that it can handle this load? Is the PCI slot PCIe 2?

Giving CPU Reservation to VM DRASTICALLY Improves Performance

geertayan — Thu, 15 Oct 2009 09:17:40 GMT

I see a similar issue regularly on several Redhat 5.2 VM guests in all our ESX clusters.

Symptoms:

Very bad performance

High CPU load in performance view in VC

Processes (such as top) running in the VM guest take up much more CPU than normal

Ping localhost gives averages of about 0.4ms (which should be around 0.04)

Fix:

Setting a small reservation (100Mhz) until ping time is normal (~0.04) again (takes up to 2 minutes)

Remove reservation

Repeat when issue happens again

I have only seen this with 2 vCPU VM's. We run patch level 123630.

Understanding Performance

drummonds — Fri, 16 May 2008 21:30:03 GMT

The following documents will explain some of the principles for virtual system performance. Please check back as we grow the number of articles here with time.

ESX and Guest Operating Systems

CPU and Scheduling

Memory

Large Memory Pages

Network

Advanced Networking Performance Options

Storage

ESX Network configuration samples

jpdicicco — Tue, 13 Oct 2009 21:58:33 GMT

Flyinverted,
I like your config except for one thing: you are using both onboard ports on 1 vswitch. You should use 1 onboard and 1 PCI vmnic on the same switch for redundancy across controllers as well. It seems like a small issue until you lose both onboard ports...

JP

How to check total Ram in the ESX host

a2alpha — Mon, 12 Oct 2009 16:01:30 GMT

I had a similar requirement a while back to find out what the configuration of the RAM and total amounts were. The following worked quite well for me:

dmidecode | grep "Size"

You'll get something like this:

Size: 4096 MB
Size: 4096 MB
Size: 4096 MB
Size: 4096 MB
Size: No Module Installed
Size: No Module Installed
Size: No Module Installed
Size: No Module Installed

Therefore 4 x 4096 = 16GB

Cas Latancy and Vmmware VMs

TobiasKracht — Mon, 12 Oct 2009 09:34:50 GMT

If you are not perfomance geek, or not using high perfomance and hign availability database with with a lot of requests - it`s not.

StarWind Software R&D

Horrible performance after long inactivity

jkounis — Sat, 10 Oct 2009 20:48:47 GMT

I am running a Windows XP (32-bit) guest on an Ubuntu 9.04 (64-bit) host using VMPlayer 2.5.3. Performance was marginal until I set "sched.mem.pshare.enable" and "mainMem.useNamedFile" to FALSE. Now performance is very good with one major exception:

If I leave the guest running overnight, when I start using it again n the morning, the performance is absolutely horrible until I reboot the guest. This morning, from clicking "Start" to clicking "Turn off this computer" was 45 sec. It took about 15 min to reboot the guest.

Is there a memory leak in VMPlayer or some other issue that causes performance to degrade over time? Is there an issue with suspend on the guest that causes somethig similar (maybe WIndows XP Power Saving or Suspend features)?

The host has 6GB of memory. I have 2GB allocated to the guest. The pertinetn information is below:

memsize = "2048"
MemAllowAutoScaleDown = "FALSE"
sched.mem.pshare.enable = "FALSE"
mainMem.useNamedFile = "FALSE"

performance problems and vmi soon depricated?

jmattson — Thu, 08 Oct 2009 15:22:38 GMT

This may help: Performance Evaluation of Intel EPT Hardware Assist.

How to handle Vendor\Application requirements (physical) in VMWare

gbattiston — Wed, 07 Oct 2009 18:38:36 GMT

thanks for the response but that's not what i'm after. There has to be some guideline to translate physical server requirements to virtual requirements. If not, then every vendor asking for 4 CPU and 8 GB RAM will be insisting on VMs built with those stats. that can't possibly be VMWare's acceptale practice.

ballooned performance counter

drummonds — Wed, 07 Oct 2009 16:20:04 GMT

The balloon driver is implemented as a device driver in the guest OS. As it inflates it takes memory from the guest and pins that memory so the guest cannot reclaim or use it. The memory is still visible to the guest, but unavailable.

By setting a memory limit to the VM you have asked the kernel to reduce the VM's memory footprint through ballooning or swapping. Because ballooning is so much more efficient that swapping, the balloon driver inflates first.

To understand the fluctuations, you would need to check swap activity on the same VM and ballooning and swap activity on other VMs. All of these activities are interrelated.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

VMWare performance resources

NTurnbull — Wed, 07 Oct 2009 15:02:17 GMT

Oh and just saw the last bit of your post again, it's not just degredation, some apps/workload benifit from being virtualised. I'd be a bit more specific on apps as there are loads of people on the communities that would be able to lend a hand and some real world experience - not just cold hard lab tests

Thanks,
Neil

Per-VM I/O Monitoring On NFS Datastores

deltajoka — Wed, 07 Oct 2009 12:07:16 GMT

Thanks for the replies!

Good to know that VMware is aware of the issue and seem to be working on a solution at least. Too bad that it seems so far away though.

I've been able to trace the file handlers used in NFS Lookup request/replies to specific files/directories (and thus determine the VMs). I've managed to filter out packets matching these file handlers with tshark in a way that i can see (and measure) the specific traffic when a VM powers on/off. However, further operations to the "disk" in the OS of the VMs don't use these file handlers. I can't trace the actual I/O, because most of that traffic is NFS FILE_SYNC packages (which have no human readable payload displaying file or directory, like the lookup packages have) and they use different file handlers than the NFS lookup packages. Too bad. Maybe there are some NFS gurus around that can nudge me in the right direction? Or suggestions about other possible workarounds?

.

Scissor — Tue, 06 Oct 2009 05:50:28 GMT

Slow RAM performance- is this normal?

Rubeck — Tue, 06 Oct 2009 07:12:47 GMT

Hi..

As this is a NUMA system, I would verify that Node Interleaving is disabled in the BIOS... Just to make sure that there's no VMs being assigned memory on a different node than the CPU it's running on.. If this happens there might be performance degradition in the VM. Even though ESX is NUMA aware and will try to CPU and mem on the same node, you never know.

The number of VMs dosn't indicate the actual number of physical CPUs in the system..
Don't know what version you're running, but in vSphere4, as an example, you can have 20 vCPUs per physical core according to docs.

/Rubeck

esxtop reports wrong zero values for CPU and Memory (Service Console)

Timo1970 — Mon, 05 Oct 2009 17:29:52 GMT

Hi!

Yes i have already read the complete document and used it while the implementation. The interesting thing is

CCPU Context Switches has correct values but

CCPU Idle Time / Usertime / Systemtime and Waittime have zero values which seems to be wrong.The

esxtop interactively shows us values for these parameters! Next to this it is the Disk performance data

(not memory ... memory values are ok! My mistake!). CMDs/a READS/s WRITE/s MBREAD/s and

MBWRTN/s report zero values all the time?

Any further idea ... ?

Kind regards ...

Timo

Any Performance Advantages Using Multiple Virtual Hard Disks? (eg, SQL Server)

DonJuanJovi — Mon, 05 Oct 2009 16:57:02 GMT

Thanks for the replies everyone. Almost there.... What about when you're VMDK's are on NFS shares (like ours!)?

CPU Utilization More

nanev — Fri, 02 Oct 2009 10:41:31 GMT

Can you give additional information:

SAS controller

number and types of HDD

types of RAID areas

logs from RAID controller (what process is going on?)

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

VSP4, VTSP4, VCP4
ITIL v.3 Foundation

Any known ways to increase Video Ram?

sflanders — Thu, 01 Oct 2009 23:36:56 GMT

I do not believe this is possible, but out of curiosity, why do you want/need to do this?

Disk write speed is very slow with Dell SAS 6/iR controller

LucasAlbers — Thu, 01 Oct 2009 16:28:26 GMT

I have a 610 and by default it has a battery back cache.

You have to specifically tell it to run with a writeback cache if you don't have a battry.

See the dell documentation.

http://www.dell.com/content/topics/topic.aspx/global/products/pvaul/topics/en/us/raid_controller?c=us&l=en&cs=555

Stress and Volume Testing

weinstein5 — Thu, 01 Oct 2009 13:07:57 GMT

Yes SVT could adversely affect the performance of other VMs depending on how your environment is designed - remember the VMs are sharing the resources with the VM/application being tested - so for example if you have a saturated a path storage the added load of the testing will affect the other VMs ability to access the storage -

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

Best hardware solution for moving our envoirnment to VM

krowczynski — Thu, 01 Oct 2009 11:21:05 GMT

Ok sou you can have 2 QUadCOre CPU with e.g. 32 GB RAM with an RAID1 with two SAS Drives, if you want to use ESX4.

If you want to use ESXi you can alos take an USB as the boot Device.

But you hardware should be supported by vmware.

http://www.vmware.com/resources/compatibility/search.php?source=hp&q=vmware%20hcl&meta=&aq=0&oq=

MCP, VCP

VM with W2003 64bits & 4GB of RAM showing less (3.75GB)

avasudevan — Thu, 01 Oct 2009 01:41:47 GMT

This is most likely due to RAM reservation for virtual PCI devices. PCI devices need memory for their IO and those pages are not available to the OS. Hence the gap that you see...

Capacity Planning and scaling with over subscription in mind..

Schorschi — Wed, 30 Sep 2009 18:57:41 GMT

However, if a given VM spikes demand for memory, beyond expectation, so the balloon process is pinched, or worse VMs are allocated memory beyond physical memory and physical memory is exceeded by demand requests, ESX host will page to disk via the swap configuration, and then every single VM, regardless of or if they requested memory will be impacted because writting to swap is painful to performance. WIth memory so inexpensive and many if not all servers running lots of RAM for virtualization, any overcommitment of memory should be avoided. Best practice is to not over subscribe memory unless you really know when and how applications in VM memory demand trends proceed. Memory overcommitment can lead to network or disk IO overcommitment beyond typical peak scaling. I have seen ESX hosts chase their respective tails trying to perform when memory oversubscription leads to network and/or disk IO additional over subscription... all because some PM or purchasing resource did not want to spend on relatively inexpensive RAM? The first thing I encourse clients to do... is increase RAM when an ESX host is struggling with this issue. You would be surprised with the results!

Virtual machines poor performance after idle time.

s1xth — Wed, 30 Sep 2009 12:11:45 GMT

Have you checked to make sure the vm's power settings are correct? No standby, no hard disk power management etc?? If the hard disks are set to turn off after a certain time that can cause this, same with any processor or system power management settings.

Linux VM, rsync throughput question

oznomad — Tue, 29 Sep 2009 23:10:14 GMT

Hiya

Thanks for that.

Finally got to the bottom of this :

The Host has 8GB of memory.
The VM for what every reason had 16GB of memory allocated.. This is not a problem, as under normal operation it doesn't ever need this.

So, when I ran the rsync on the VM, linux kindly allocated 14G to disk cache to help with rsync. The HOST doesn't know what the VM is using the memory for, so tried to service the request for the extra memory. Not having the memory, it started to to use SWAP space. This is why rsync ran at full wack for a while and then slowed once it started to use SWAP.

I only noticed this when the drop in speed seem to coincide with the request fro SWAP.

Solution : I dropped the memory on the VM to 4GB..to stop the host using SWAP, rsync was happy.

Hope this help

Kind Regards

nomad

VM CPU issue?

drummonds — Tue, 29 Sep 2009 02:07:11 GMT

The uniprocessor HAL is more efficient than the multiprocessor HAL. If you only have a single vCPU assigned to your VM, you can improve performance slightly by forcing Windows back to the uniprocessor HAL. You can do this through the device manager and by specifying a driver to the CPU.

The performance difference generally is not great, though. We have measured it in single digit percentages that are workload dependent.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

XP guest slows down when moved to Quad / Vista 64 from XP host

drummonds — Tue, 29 Sep 2009 02:04:26 GMT

The only lead I have for you is around Intel's SpeedStep technology. SpeedStep will increase/decrease the processor frequency based on guest load. A few months ago I discovered an issue in our hosted products where load inside a VM was not detected by SpeedStep, so the processor ran at very low frequencies. Check your BIOS and SpeedStep settings (if available) for "max performance" and see if this makes a difference.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

Virtulize SQL on local storage

drummonds — Tue, 29 Sep 2009 01:59:35 GMT

The issue here isn't to use ESX or not, but whether you run SQL on the servers available to you. The overhead of SQL virtualization is so low with ESX 4, and the IO throughput so high, that ESX's presence will be transparent to the application and its users. You simply must estimate your transaction throughput and subsequent IOPS and assure that your storage can support such a rate.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

Quad core - and 32 bit application

Troy Clavell — Sun, 27 Sep 2009 14:02:38 GMT

download and install ESXi, it will give you the ability to assign 4vCPU's to your guest(s) if needed.
http://www.vmware.com/download/esxi/

....but check the VMware HCL or the Whitebox HCL

vCenter Performance Counters

drummonds — Fri, 30 May 2008 00:15:21 GMT

Introduction

The following table of vCenter (VC) performance counters lists the counters with a description of their purpose. This page has been updated for vSphere 4, so the counter levels will differ slightly on older versions of VC.

Remember, with the exception of ready time, statistic levels one and two are the only ones needed for 99% of the performance monitoring and analysis out there. Don't spend many of your own cycles worrying about levels three and four!

For information on enabling VC to display and archive these counters see the Understanding vCenter Performance Statistics article.

Understanding vCenter Measurement Windows

Before you continue, you should know that all total count metrics reported by VC are reported over the sample window. When you're looking at live stats, this sample window is 20 seconds. When you're looking at archive stats, it will depend on the interval duration. That duration could be five minutes, 30 minutes, two hours, or one day.

This causes a lot of confusion when comparing esxtop results to live VC results to archived VC results. As an example, ready time might be reported as 10% in esxtop. In live VC results this amount of ready time would be reported as 2000 ms (10% of the 20s window.) In one day archive results, the same number would be reported as 30,000 ms (10% of the five minute interval duration.) All of these numbes reflect the same amount of ready time.

CPU Statistics

Level	Counter name in API	Description	Units
1	cpu.ready.summation	Ready time is the time spend waiting for CPU(s) to become available in the past update interval.	millisecond
1	cpu.usagemhz.average	The CPU utilization. The maximum possible value here is the frequency of the processors times the number of cores. As an example, a VM using 4000 MHz on a system with four 2 GHz processors is using 50% of the CPU (4000 / (4 * 2000) = 0.5)	megaHertz
1	cpu.usage.average	The CPU utilization. This value is reported with 100% representing all processor cores on the system. As an example, a 2-way VM using 50% of a four-core system is completely using two cores.	percent
2	cpu.reservedCapacity.average	CPU Reserved Capacity	megaHertz
2	cpu.idle.summation	CPU Idle	millisecond
2	cpu.swapwait.summation	Swap wait time is time that the world spent waiting for memory to be swapped in. When the VM is waiting for memory, it is not doing work.	millisecond
3	cpu.system.summation	System time is the time spent in VMkernel during the last update interval. This does not include guest code execution.	millisecond
3	cpu.wait.summation	Wait time is the time spent waiting for hardware or VMkernel lock thread locks during the last update interval.	millisecond
3	cpu.extra.summation	CPU extra is the time above the statically calculated entitlement. Entitlement is the share of processing time that a VM should get as a result of its vCPU count and assigned shares. You should not use or care about this counter in any of your own analysis.	millisecond
3	cpu.used.summation	CPU Used	millisecond
3	cpu.guaranteed.latest	Guaranteed time is reported as the amount of the reservation time that the VM used in the past update interval. As an example, if 2000 MHz have been reserved for the VM on an four-way, 2 GHz host, that's 25% of the CPU resource. In a 20s update interval, there are 80,000 ms available on this four-way system. That means 20,000 ms of time has been reserved. If a VM used only half of its available cycles, the guaranteed time is 10,000 ms.	millisecond
4	cpu.usage.none	CPU Usage (None)	percent
4	cpu.usage.minimum	CPU Usage (Minimum)	percent
4	cpu.usage.maximum	CPU Usage (Maximum)	percent
4	cpu.usagemhz.none	CPU Usage in MHz (None)	megaHertz
4	cpu.usagemhz.minimum	CPU Usage in MHz (Minimum)	megaHertz
4	cpu.usagemhz.maximum	CPU Usage in MHz (Maximum)	megaHertz

Memory Statistics

Level	Counter name in API	Description	units
1	mem.consumed.average	The amount of machine memory that is in use by the VM. While a VM may have been configured to use 4 GB of RAM, as an example, it might have only touched half of that. Of the 2 GB left, half of that might be saved from memory sharing. That would result in 1 GB of consumed memory.	kiloBytes
1	mem.overhead.average	The memory used by the VMkernel to maintain and execute the VM.	kiloBytes
1	mem.swapinrate.average	The swap in rate reports the rate at which a VM's memory is being swapped in from disk.	kiloBytesPerSecond
1	mem.swapoutrate.average	The swap out rate reports the rate at which a VM's memory is being swapped out to disk.	kiloBytesPerSecond
1	mem.usage.average	The percentage of memory used as a percent of all available machine memory. Available for host and VM.	percent
1	mem.vmmemctl.average	The amount of memory currently claimed by the balloon driver. This is not a performance problem, per se, but represents the host starting to take memory from less needful VMs for those with large amounts of active memory. But if the host is ballooning, check swap rates (swapin and swapout) which would be indicative of performance problems.	kiloBytes
2	mem.granted.average	The amount of memory that was granted to the VM by the host. Memory is not granted to the host until it is touched one time and granted memory may be swapped out or ballooned away if the VMkernel needs the memory.	kiloBytes
2	mem.active.average	The amount of memory used by the VM in the past small window of time. This is the "true" number of how much memory the VM currently has need of. Additional, unused memory may be swapped out or ballooned with no impact to the guest's performance.	kiloBytes
2	mem.shared.average	The average amount of shared memory. Shared memory represents the entire pool of memory from which sharing savings are possible. The amount of memory that this has been condensed to is reported in shared common memory. So, total saving due to memory sharing equals shared memory minus shared common memory.	kiloBytes
2	mem.zero.average	The amount of zero pages in the guest. Zero pages are not represented in machine memory so this results in 100% savings when mapping from the guest to the machine memory.	kiloBytes
2	mem.unreserved.average	Memory Unreserved (Average)	kiloBytes
2	mem.swapused.average	The amount of swap memory currently in use. A large amount of swap memory is not a performance problem. This could be memory that the guest doesn't need. Check the swap rates (swapin, swapout) to see if the guest is actively in need of more memory than is available.	kiloBytes
2	mem.swapunreserved.average	Memory Swap Unreserved (Average)	kiloBytes
2	mem.sharedcommon.average	The average amount of shared common memory. Shared memory represents the entire pool of memory from which sharing savings are possible. The amount of memory that this has been condensed to is reported in shared common memory. So, total saving due to memory sharing equals shared memory minus shared common memory.	kiloBytes
2	mem.heap.average	Memory Heap (Average)	kiloBytes
2	mem.heapfree.average	Memory Heap Free (Average)	kiloBytes
2	mem.state.latest	Memory State	number
2	mem.swapped.average	Memory Swapped (Average)	kiloBytes
2	mem.swaptarget.average	Memory Swap Target (Average)	kiloBytes
2	mem.swapin.average	The rate at which memory is being swapped in from disk. A large number here represents a problem with lack of memory and a clear indication that performance is suffering as a result.	kiloBytes
2	mem.swapout.average	The rate at which memory is being swapped out to disk. A large number here represents a problem with lack of memory and a clear indication that performance is suffering as a result.	kiloBytes
2	mem.vmmemctltarget.average	Memory Balloon Target (Average)	kiloBytes
2	mem.sysUsage.average	Memory Used by vmkernel	kiloBytes
2	mem.reservedCapacity.average	Memory Reserved Capacity	megaBytes
4	mem.usage.none	Memory Usage (None)	percent
4	mem.usage.minimum	Memory Usage (Minimum)	percent
4	mem.usage.maximum	Memory Usage (Maximum)	percent
4	mem.granted.none	Memory Granted (None)	kiloBytes
4	mem.granted.minimum	Memory Granted (Minimum)	kiloBytes
4	mem.granted.maximum	Memory Granted (Maximum)	kiloBytes
4	mem.active.none	Memory Active (None)	kiloBytes
4	mem.active.minimum	Memory Active (Minimum)	kiloBytes
4	mem.active.maximum	Memory Active (Maximum)	kiloBytes
4	mem.shared.none	Memory Shared (None)	kiloBytes
4	mem.shared.minimum	Memory Shared (Minimum)	kiloBytes
4	mem.shared.maximum	Memory Shared (Maximum)	kiloBytes
4	mem.zero.none	Memory Zero (None)	kiloBytes
4	mem.zero.minimum	Memory Zero (Minimum)	kiloBytes
4	mem.zero.maximum	Memory Zero (Maximum)	kiloBytes
4	mem.unreserved.none	Memory Unreserved (None)	kiloBytes
4	mem.unreserved.minimum	Memory Unreserved (Minimum)	kiloBytes
4	mem.unreserved.maximum	Memory Unreserved (Maximum)	kiloBytes
4	mem.swapused.none	Memory Swap Used (None)	kiloBytes
4	mem.swapused.minimum	Memory Swap Used (Minimum)	kiloBytes
4	mem.swapused.maximum	Memory Swap Used (Maximum)	kiloBytes
4	mem.swapunreserved.none	Memory Swap Unreserved (None)	kiloBytes
4	mem.swapunreserved.minimum	Memory Swap Unreserved (Minimum)	kiloBytes
4	mem.swapunreserved.maximum	Memory Swap Unreserved (Maximum)	kiloBytes
4	mem.sharedcommon.none	Memory Shared Common (None)	kiloBytes
4	mem.sharedcommon.minimum	Memory Shared Common (Minimum)	kiloBytes
4	mem.sharedcommon.maximum	Memory Shared Common (Maximum)	kiloBytes
4	mem.heap.none	Memory Heap (None)	kiloBytes
4	mem.heap.minimum	Memory Heap (Minimum)	kiloBytes
4	mem.heap.maximum	Memory Heap (Maximum)	kiloBytes
4	mem.heapfree.none	Memory Heap Free (None)	kiloBytes
4	mem.heapfree.minimum	Memory Heap Free (Minimum)	kiloBytes
4	mem.heapfree.maximum	Memory Heap Free (Maximum)	kiloBytes
4	mem.swapped.none	Memory Swapped (None)	kiloBytes
4	mem.swapped.minimum	Memory Swapped (Minimum)	kiloBytes
4	mem.swapped.maximum	Memory Swapped (Maximum)	kiloBytes
4	mem.swaptarget.none	Memory Swap Target (None)	kiloBytes
4	mem.swaptarget.minimum	Memory Swap Target (Minimum)	kiloBytes
4	mem.swaptarget.maximum	Memory Swap Target (Maximum)	kiloBytes
4	mem.swapin.none	Memory Swap In (None)	kiloBytes
4	mem.swapin.minimum	Memory Swap In (Minimum)	kiloBytes
4	mem.swapin.maximum	Memory Swap In (Maximum)	kiloBytes
4	mem.swapout.none	Memory Swap Out (None)	kiloBytes
4	mem.swapout.minimum	Memory Swap Out (Minimum)	kiloBytes
4	mem.swapout.maximum	Memory Swap Out (Maximum)	kiloBytes
4	mem.vmmemctl.none	Memory Balloon (None)	kiloBytes
4	mem.vmmemctl.minimum	Memory Balloon (Minimum)	kiloBytes
4	mem.vmmemctl.maximum	Memory Balloon (Maximum)	kiloBytes
4	mem.vmmemctltarget.none	Memory Balloon Target (None)	kiloBytes
4	mem.vmmemctltarget.minimum	Memory Balloon Target (Minimum)	kiloBytes
4	mem.vmmemctltarget.maximum	Memory Balloon Target (Maximum)	kiloBytes
4	mem.overhead.none	Memory Overhead (None)	kiloBytes
4	mem.overhead.minimum	Memory Overhead (Minimum)	kiloBytes
4	mem.overhead.maximum	Memory Overhead (Maximum)	kiloBytes
4	mem.consumed.none	Memory Consumed (None)	kiloBytes
4	mem.consumed.maximum	Memory Consumed (Maximum)	kiloBytes
4	mem.consumed.minimum	Memory Consumed (Minimum)	kiloBytes
4	mem.sysUsage.none	Memory Used by vmkernel	kiloBytes
4	mem.sysUsage.maximum	Memory Used by vmkernel	kiloBytes
4	mem.sysUsage.minimum	Memory Used by vmkernel	kiloBytes

Disk Statistics

Level	Counter name in API	Description	units
1	disk.maxTotalLatency	The highest reported total latency (device and kernel times) in the sample window.	milliseconds
1	disk.usage.average	Average disk throughput over the sample period.	kiloBytesPerSecond
2	disk.read.average	Average disk throughput due to read operaitons over the sample period.	kiloBytesPerSecond
2	disk.write.average	Average disk throughput due to write operations over the sample period.	kiloBytesPerSecond
2	disk.commands.summation	Disk Commands Issued	number
2	disk.commandsAborted.summation	The number of aborts that have occurred in the last window of time. Abort commands are issued by the guest when the storage system has not responded within an acceptable amount of time (as defined by the guest OS or application.)	number
2	disk.busResets.summation	Disk Bus Resets	number
2	disk.deviceReadLatency.average	Device read latency. This is the time the physical device from the HBA to the platter takes to service an IO request.	millisecond
2	disk.kernelReadLatency.average	Kernel read latency. This is the time the VMkernel takes to service an IO. This is the time between the guest OS and the device.	millisecond
2	disk.totalReadLatency.average	Total read latency. The sum of the device and kernel read latencies.	millisecond
2	disk.queueReadLatency.average	Queue Read Latency	millisecond
2	disk.deviceWriteLatency.average	Device write latency. This is the time the physical device from the HBA to the platter takes to service an IO request.	millisecond
2	disk.kernelWriteLatency.average	Kernel write latency. This is the time the VMkernel takes to service an IO. This is the time between the guest OS and the device.	millisecond
2	disk.totalWriteLatency.average	Total write latency. The sum of the device and kernel write latencies.	millisecond
2	disk.queueWriteLatency.average	Queue Write Latency	millisecond
2	disk.deviceLatency.average	Physical Device Command Latency	millisecond
2	disk.kernelLatency.average	Kernel Disk Command Latency	millisecond
2	disk.queueLatency.average	Queue Command Latency	millisecond
3	disk.numberRead.summation	The number of IO read operations in the previous sample period. Note that these operations may be variable sized up to 64 KB.	number
3	disk.numberWrite.summation	The number of IO write operations in the previous sample period. Note that these operations may be variable sized up to 64 KB.	number
3	disk.totalLatency.average	This is the average total latency over the sample window. Total latency is the sum of kernel and device latency for both read and write commands.	millisecond
3	disk.write.average	Disk Write Rate	kiloBytesPerSecond
4	disk.usage.none	Disk Usage (None)	kiloBytesPerSecond
4	disk.usage.minimum	Disk Usage (Minimum)	kiloBytesPerSecond
4	disk.usage.maximum	Disk Usage (Maximum)	kiloBytesPerSecond

Network Statistics

Level	Counter name in API	Description	units
1	net.usage.average	Network Usage (Average)	kiloBytesPerSecond
2	net.droppedRx.summation	The number of received packets that were dropped over the sample period.	number
2	net.droppedTx.summation	The number of transmitted packets that were dropped over the sample period.	number
2	net.received.average	Average network throughput for received traffic.	kiloBytesPerSecond
2	net.transmitted.average	Average network throughput for transmitted traffic.	kiloBytesPerSecond
3	net.packetsRx.summation	Network Packets Received	number
3	net.packetsTx.summation	Network Packets Transmitted	number
4	net.usage.none	Network Usage (None)	kiloBytesPerSecond
4	net.usage.minimum	Network Usage (Minimum)	kiloBytesPerSecond
4	net.usage.maximum	Network Usage (Maximum)	kiloBytesPerSecond

Other Statistics

Level	Counter name in API	Description	units
1	sys.uptime.latest	Uptime	second
1	sys.heartbeat.summation	Heartbeat	number
1	clusterServices.cpufairness.latest	CPU Fairness	number
1	clusterServices.memfairness.latest	Memory Fairness	number
1	clusterServices.effectivecpu.average	Effective CPU Resources	megaHertz
1	clusterServices.effectivemem.average	Effective Memory Resources	megaBytes
1	clusterServices.failover.latest	Current failover level	number
3	sys.resourceCpuUsage.average	Resource CPU Usage (Average)	megaHertz
3	managementAgent.memUsed.average	Memory Used (Average)	kiloBytes
3	managementAgent.swapUsed.average	Memory Swap Used (Average)	kiloBytes
3	managementAgent.swapIn.average	Memory Swap In (Average)	kiloBytesPerSecond
3	managementAgent.swapOut.average	Memory Swap Out (Average)	kiloBytesPerSecond
3	rescpu.actav1.latest	CPU Active (1 min. average)	percent
3	rescpu.actpk1.latest	CPU Active (1 min. peak)	percent
3	rescpu.runav1.latest	CPU Running (1 min. average)	percent
3	rescpu.actav5.latest	CPU Active (5 min. average)	percent
3	rescpu.actpk5.latest	CPU Active (5 min. peak)	percent
3	rescpu.runav5.latest	CPU Running (5 min. average)	percent
3	rescpu.actav15.latest	CPU Active (15 min. average)	percent
3	rescpu.actpk15.latest	CPU Active (15 min. peak)	percent
3	rescpu.runav15.latest	CPU Running (15 min. average)	percent
3	rescpu.runpk1.latest	CPU Running (1 min. peak)	percent
3	rescpu.maxLimited1.latest	CPU Throttled (1 min. average)	percent
3	rescpu.runpk5.latest	CPU Running (5 min. peak)	percent
3	rescpu.maxLimited5.latest	CPU Throttled (5 min. average)	percent
3	rescpu.runpk15.latest	CPU Running (15 min. peak)	percent
3	rescpu.maxLimited15.latest	CPU Throttled (15 min. average)	percent
3	rescpu.sampleCount.latest	Group CPU Sample Count	number
3	rescpu.samplePeriod.latest	Group CPU Sample Period	millisecond
4	sys.resourceCpuUsage.none	Resource CPU Usage (None)	megaHertz
4	sys.resourceCpuUsage.maximum	Resource CPU Usage (Maximum)	megaHertz
4	sys.resourceCpuUsage.minimum	Resource CPU Usage (Minimum)	megaHertz

Packet loss on Sun Fire X4140 / ESXi 3.5u4

HMVM123 — Thu, 24 Sep 2009 19:18:20 GMT

No, I couldn't make much progress. Just to be clear, the NICs were functional and my observations were related to packet loss at high throughputs.

I had the SunFire under an eval program, so I had to return it. I'm now using an IBM x3550M2 and am not seeing this. Have also upgraded to ESXi4.0 and am using VmxNet drivers on the interfaces.

SQl server performance down after migration from P to V..now we have allocated 6 GB ram and 4vCPU and 3 GB reserverd RAM but no improvement..its taking double time for mdf data copying in copriason with P

Chuck8773 — Wed, 23 Sep 2009 18:12:43 GMT

How much RAM is in the ESX Host? How many CPU cores are in the host? How many other VM's are on that host? What kind of disk is the VM on, SATA, SAS, FC?

Charles Killmer, VCP

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

Virtual Machine Swap file location

tom howarth — Tue, 22 Sep 2009 21:24:26 GMT

Another good reason for centalising your Swap is so that you can remove it from a backup strategy.

If you found this or any other answer useful please consider the use of the Helpful or correct buttons to award points

Tom Howarth VCP / vExpert
VMware Communities User Moderator
Blog: www.planetvm.net
Contributing author for the upcoming book "VMware vSphere and Virtual Infrastructure Security: Securing ESX and the Virtual Environment”. Currently available on roughcuts

Memory Allocation Alert raised yet VM is well below 50%. VC is reporting wrong

armaniseal — Tue, 22 Sep 2009 13:09:32 GMT

Thanks ... that explains it.. its a general problem. Hope they fix it soon because alerts are useless like this.

VMware ESX 3.02, c-Class Enclosures & HP EVA 8x00

VCP210708 — Mon, 21 Sep 2009 19:32:07 GMT

Hello,

Im hoping someone will be able to assist me with regards to read and write latency.

I have been doing some baseline storage reports for the EXA's that host our VMware Virtual Desktop Infrastructure.

The setup I have is HP c-Class enclosures running VMware. 2 x 8 Server clusters per enclosure. Connectivity and Server hardware are BL485 G5 Servers - HBA - mds9124e internal - mds9124e external - (NPV mode) - Core Switch - Storage Port (EVA) - EVA 8100.

Load is evenly distributed across all storage processor ports which the Controller Status Performance Object is showing. 1 controller is working slightly more as the BFS volumes all go through Port 1 on Controller A due to the installation LUN must be the lowest-numbered of the LUNs available to the physical host.

Do we have any figures as to limitations on the following EVAPerf counters?

Host Port Statistics Performance Object
Read Req/s & Write Req/s
Read MB/s & Write MB/s
Read Latency & Write Latency
Av Queue Depth

Queue Depth Readings are around 1-2 which looks to be fine, host port Stats look to be fine with utilization on the paired controllers below 50%

The issue where I need assistance is the Average Read Latency & Average Write Latency, I understand that Aa average Read Latency < 15ms and Average Write Latency < 5 ms should be acceptable for most workloads, does this also cound for VDI's?

I am seeing the readings below, read latency is fine, however the write latency is above the 5 (ms) recomendation, now this recomendation isnt set in stone and so if there are any concrete acceptable / not acceptable figures out there supplied by HP or VMware I would appreashiate a benchmark figure.

Totals	Average Read Latency (ms)	Average Write Latency (ms)
DATA_DG1	9.117820069	8.909515571
DATA_DG2	9.11710754	8.340271941

Kind Regrads

Deano

Java Performance on VMware ESX

tcutts — Fri, 18 Sep 2009 14:45:27 GMT

Having gone through those best practices, yes, we're on board with all of them.

The only variable we have here is the 'hardware' platform on which the tomcat java service is running. The software is on an NFS server, so both physical and virtual servers are getting their tomcat copy from the same place. The same goes for the Oracle client software. The Oracle server is on a physical machine, and is the same for both the physical and virtual clients. The OS installed on the physical and virtual machines is also identical, with the exception of some of the device driver modules in use, a natural consequence of virtualising the system.

So we have already controlled for most variables. I'm still waiting for the broken-down code snippets from the users.

Tim

Reservations?? resource priorites??

wade001 — Sun, 20 Sep 2009 13:09:07 GMT

Thanks much. You comments were very helpful.

IBM x3850 M2 VM performance issues?

vikrams04 — Sat, 19 Sep 2009 13:38:18 GMT

Hi,

There's a similar issue faced by me in one of my deployments. Even after upgrading to the latest BMC Update given by IBM on its site for

X3850 M2 has not given us positive results.

The issue is specifically with I/O intensive Servers like Citrix Xe App 4.5 where it touches 100 % CPU after having 15-20 Users.

Insufficient system resources when running multiple vm

drummonds — Fri, 18 Sep 2009 16:59:30 GMT

I am going to assume that when you say "images" you are referring to virtual machines.

ESX (indeed, all of our virtualization products) provide hardened, isolated virtual machines that are not aware of anything else happening on the system. So, no matter how many VMs you instantiate, the guest OSes in each will not be aware of memory pressure in the host.

There is one exception to this: once memory has been over-committed the balloon driver will inflate inside the guest, pinning memory, and prohibiting the guest OS and its applications from using that memory. This could explain what you are seeing. But, since the balloon driver's maximum size is limited to 65% of the VM's size, balloon inflation should not cause guest OS or application crashes like you describe.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

Help to understand how to monitord esx with third part tool

drummonds — Fri, 18 Sep 2009 16:45:46 GMT

I do not have any experience with third-party tools but I know that they are dependent on the SDK, which provides access to the same counters that vCenter uses. I have documented VC's counters (vCenter Performance Counters) and provided some documentation as to how to use them (Performance Monitoring and Analysis). Try reading those.

Scott

More information on my blog and on Twitter:
http://vpivot.com
http://twitter.com/drummonds

Performance Troubleshooting for VMware vSphere 4 and ESX 4.0

haroldr — Mon, 13 Jul 2009 14:03:44 GMT

Performance problems can arise in any computing environment. Complex application behaviors, changing demands, and shared infrastructure can lead to problems arising in previously stable environments. Troubleshooting performance problems requires an understanding of the interactions between the software and hardware components of a computing environment. Moving to a virtualized computing environment adds new software layers and new types of interactions that must be considered when troubleshooting performance problems.

The attached document is the first installment in a guide covering performance troubleshooting in a vSphere environment. It uses a guided approach to lead the reader through the observable manifestations of complex hardware/software interactions in order to identify specific performance problems. For each problem covered, it includes a discussion of the possible root-causes and solutions. Topics covered include performance problems arising from issues in the CPU, memory, storage, and network subsystems, as well as in the VM and ESX host configuration. Guidance is given on relevant performance metrics to observe using the vSphere Client and esxtop in order to isolate specific performance issues.

This first installment of Performance Troubleshooting for VMware vSphere 4 covers performance troubleshooting on a single VMware ESX 4.0 host. It focuses on the most common performance problems which affect an ESX host. Future updates will add more detailed performance information, including troubleshooting information for more advanced problems and multi-host vSphere deployments.

This is a living document. Reader comments, questions, and suggestions are encouraged.

how long does the vminstaller take to boot

cmlarson — Thu, 17 Sep 2009 19:43:05 GMT

the cpu's on this server do not support the 64 bit vmware. I must have confused 3.5 update 4 with esxi 4.0. so, i'm installing the 3.5 version

Intel E5530 vs Intel L5530

LucasAlbers — Thu, 17 Sep 2009 17:29:34 GMT

It appears from that chart that their are a number of differences, more than just turbo boost.

Intel® Virtualization Technology
Execute Disable Bit
Enhanced Intel® Speedstep Technology
Enhanced Halt State (C1E)
Intel® 64
Intel® Demand Based Switching
Intel® Turbo Boost Technology
Intel® Hyper-Threading Technology
Intel® Trusted Execution Technology
Intel® Virtualization Technology for Directed I/O