VMware Communities: Message List - Open unofficial storage performance thread

Open unofficial storage performance thread

christianZ — Sat, 24 Feb 2007 17:14:52 GMT

Attention!
Since this thread is getting longer and longer, not to mention the load times, Christian and I decided to close this thread and start a new one.

The new thread is available here: New !! Open unofficial storage performance thread

Oliver Reeh
VMware Communities User Moderator

My idea is to create an open thread with uniform tests whereby the results will be all inofficial and w/o any
warranty.

If anybody shouldn't be agreed with some results then he can make own tests and presents
his/her results too.

I hope this way to classify the different systems and give a "neutral" performance comparison.

Additionally I will mention that the performance is one of many aspects to choose the right system.

The others could be e.g.
- costs for storage system and infrastructure, etc.

There are examples of IOMETER Tests:
=====================================

######## TEST NAME: Max Throughput-100%Read
size,% of size,% reads,% random,delay,burst,align,reply
32768,100,100,0,0,1,0,0

######## TEST NAME: RealLife-60%Rand-65%Read
size,% of size,% reads,% random,delay,burst,align,reply
8192,100,65,60,0,1,0,0

######## TEST NAME: Max Throughput-50%Read
size,% of size,% reads,% random,delay,burst,align,reply
32768,100,50,0,0,1,0,0

######## TEST NAME: Random-8k-70%Read
size,% of size,% reads,% random,delay,burst,align,reply
8192,100,70,100,0,1,0,0

The global options are:
=====================================

Worker
Worker 1
Worker type
DISK
Default target settings for worker
Number of outstanding IOs,test connection rate,transactions per connection
64,ENABLED,500
Disk maximum size,starting sector
8000000,0

Run time = 5 min

For testing the disk C is configured and the test file (8000000 sectors) will be created by
first running - you need free space on the disk.

The cache size has direct influence on results. By systems with cache over 2GB the test
file should be increased.

LINK TO IOMETER:
http://sourceforge.net/projects/iometer/

Significant results are: Av. Response time, Av. IOS/sek, Av. MB/s
To mention are: what server (vm or physical), Processor number/type; What storage system, How many disks

Here the config file *.icf
####################################### BEGIN of *.icf

Version 2004.07.30
'TEST SETUP ====================================================================
'Test Description
IO-Test
'Run Time
' hours minutes seconds
0 5 0
'Ramp Up Time (s)
0
'Default Disk Workers to Spawn
NUMBER_OF_CPUS
'Default Network Workers to Spawn
0
'Record Results
ALL
'Worker Cycling
' start step step type
1 5 LINEAR
'Disk Cycling
' start step step type
1 1 LINEAR
'Queue Depth Cycling
' start end step step type
8 128 2 EXPONENTIAL
'Test Type
NORMAL
'END test setup
'RESULTS DISPLAY ===============================================================
'Update Frequency,Update Type
4,WHOLE_TEST
'Bar chart 1 statistic
Total I/Os per Second
'Bar chart 2 statistic
Total MBs per Second
'Bar chart 3 statistic
Average I/O Response Time (ms)
'Bar chart 4 statistic
Maximum I/O Response Time (ms)
'Bar chart 5 statistic
% CPU Utilization (total)
'Bar chart 6 statistic
Total Error Count
'END results display
'ACCESS SPECIFICATIONS =========================================================
'Access specification name,default assignment
Max Throughput-100%Read,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
32768,100,100,0,0,1,0,0
'Access specification name,default assignment
RealLife-60%Rand-65%Read,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
8192,100,65,60,0,1,0,0
'Access specification name,default assignment
Max Throughput-50%Read,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
32768,100,50,0,0,1,0,0
'Access specification name,default assignment
Random-8k-70%Read,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
8192,100,70,100,0,1,0,0
'END access specifications
'MANAGER LIST ==================================================================
'Manager ID, manager name
1,PB-W2K3-04
'Manager network address
193.27.20.145
'Worker
Worker 1
'Worker type
DISK
'Default target settings for worker
'Number of outstanding IOs,test connection rate,transactions per connection
64,ENABLED,500
'Disk maximum size,starting sector
8000000,0
'End default target settings for worker
'Assigned access specs
'End assigned access specs
'Target assignments
'Target
C:
'Target type
DISK
'End target
'End target assignments
'End worker
'End manager
'END manager list
Version 2004.07.30

####################################### ENDE of *.icf

TABLE SAMPLE

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM or PHYS.
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell PE6850, 16GB RAM; 4x XEON 51xx, 2,66 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 1 / 14+2 Disks / R50

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........__________..........__________.........__________

RealLife-60%Rand-65%Read......__________..........__________.........__________

Max Throughput-50%Read..........__________..........__________.........__________

Random-8k-70%Read.................__________..........__________.........__________

EXCEPTIONS: CPU Util.-XX%;

##################################################################################

I hope YOU JOIN IN !

Regards
Christian

A Google Spreadsheet version is here:
http://spreadsheets.google.com/pub?key=p2IFgyUF_v5Jn-7QobgY9Fw

Message was edited by:
ken.cline@hp.com to remove ALL CAPS from thread title

Message was edited by:
RDPetruska
Added link to Atamido's Google Spreadsheet

Re: Open unofficial storage performance thread

oreeh — Thu, 05 Mar 2009 14:53:48 GMT

Since this thread is getting longer and longer, not to mention the load times, Christian and I decided to close this thread and start a new one.

The new thread is available here: New !! Open unofficial storage performance thread

Oliver Reeh
VMware Communities User Moderator

Re: Open unofficial storage performance thread

christianZ — Thu, 05 Mar 2009 13:48:48 GMT

Ok when I now compare your results with that from mnemonic and my then I would say the box works poor with R5 config - but not sure here.

Re: Open unofficial storage performance thread

Mnemonic — Thu, 05 Mar 2009 13:06:46 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: Windows 2003 std VM on ESX 3.5.0 Update 3
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: IBM x3650, 16GB RAM; 2x XEON E5410 2,33 GHz Quad-Core
SAN Type: IBM DS3400 / Disks: 68GB 15k SAS / RAID LEVEL: Raid5 / 8 Disks / QLA 2432 4Gbit FC HBA

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read.......______5___..........___11419__........___357____

RealLife-60%Rand-65%Read..._____66___.........._____726__........_____6____

Max Throughput-50%Read.........______8___..........____7666__........___240____

Random-8k-70%Read................_____78___.........._____704__........_____6____

##################################################################################

Re: Open unofficial storage performance thread

hicksj — Thu, 05 Mar 2009 13:41:01 GMT

I had posted similar numbers a while back on a DS3300. Wasn't very impressive.

http://communities.vmware.com/message/943905#943905

My response times were similar to yours christianZ, but with IO under 500 with the "reallife" test. Of course, I was testing with only 3 drives at the time. I posted non-ESX (physical hardware) specs against the DS3300 and they were consistent with the VM results.

Re: Open unofficial storage performance thread

christianZ — Thu, 05 Mar 2009 13:30:29 GMT

by me only 512 MB, the response times were better too - ca. 40 ms

Re: Open unofficial storage performance thread

Mnemonic — Thu, 05 Mar 2009 13:22:56 GMT

I must say it seams to be performing preatty poorly.. I am not sure how much cache is in the controller. I will investigate.

Re: Open unofficial storage performance thread

christianZ — Thu, 05 Mar 2009 13:12:31 GMT

Great, thanks - is that throughput enough for you? I've just tested DS3400 with 6XSAS15k as R10 and saw ca. 1800 ios/sec (by Reallife-test).

Re: Open unofficial storage performance thread

christianZ — Thu, 26 Feb 2009 07:27:09 GMT

Thanks for that.

How many disks were used in the aggregate?

What features were on (dedup, thinprovisioning, etc.)

Re: Open unofficial storage performance thread

RParker — Thu, 26 Feb 2009 00:51:40 GMT

ESX: 3.5 U3 143128
SYSTEM:IBM 3950 M2 / 64GB / Emulex Fiber SAN / Netapp 3070 / 15K SAS
VM: Windows 2003 R2 / 1GB

Access Specification Name	IOps	Read IOps	Write IOps	MBps	Read MBps	Write MBps	Average Response Time
Max Throughput-100%Read	3844.361848	3844.361848	0	120.136308	120.136308	0	5.735864

RealLife-60%Rand-65%Read	2723.538409	1772.493252	951.045157	21.277644	13.847604	7.43004	19.332422

Max Throughput-50%Read	4127.849316	2064.249641	2063.599676	128.995291	64.507801	64.48749	7.096561

Random-8k-70%Read	2155.345519	1509.023551	646.321968	16.838637	11.789246	5.04939	24.746402

Re: DS3200 ESX 3.5 vs XenServer

typera — Thu, 26 Feb 2009 00:08:15 GMT

And here are the final summary of RAID10 and RAID5 tests with ESX vs XenServer:

see attachment.

Cheers,

Max

www.intellectit.com.au

Re: DS3200 ESX 3.5 vs XenServer

typera — Wed, 25 Feb 2009 23:50:53 GMT

Further update, after realigning partitions under RAID5, its still a tie between ESX and XenServer:

Test name	ESX	XEN	ESX	XEN	ESX	XEN	ESX	XEN
Max Throughput-100%Read	2.4	3.1	20,436.3	19,227.3	638.6	600.9	81.4	19.7
RealLife-60%Rand-65%Read	152.4	153.6	354.4	359.2	2.8	2.8	23.6	21.1
Max Throughput-50%Read	7.1	9.1	8,385.1	6,582.3	262.0	205.7	41.1	13.2
Random-8k-70%Read	165.7	166.2	309.9	314.3	2.4	2.5	28.3	26.0

Re: DS3200 ESX 3.5 vs XenServer

typera — Wed, 25 Feb 2009 13:35:13 GMT

Just redone the benchmarks under RAID 5 (5 x 750Gb with hot spare):

Based on these results ESX wins for Avg Response times and Xen for CPU, its a draw for IOs and throughput.

Whats interesting though, is that under RAID10 XEN peforms better, wins for pretty much all counters except for 100% read (except cpu)...

Anyone have any idea how Xen can be so unaffected by LUN level block size?

Server	IBM x3650	RAM	28 GB (667)
CPU	2 x Intel X5450	SAN	IBM DS3200 (SAS)
Num Drives	6 x 750GB SATA	RAID Level	RAID5 + 1 Hot Spare

VMWare ESX 3.5 Update 3		Citrix XenServer Enterprise 5.0.0

SAN Cache	1024MB	SCSI Controller	LSI		SAN Cache	1024MB
Block (Segment) size	512	VMFS block size	1MB		Block (Segment) size	512
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation	Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	2.4	20,170.9	630.3	84.1	Max Throughput-100%Read	3.1	19,227.3	600.9	19.7
RealLife-60%Rand-65%Read	149.3	357.8	2.8	24.4	RealLife-60%Rand-65%Read	153.6	359.2	2.8	21.1
Max Throughput-50%Read	8.8	6,665.6	208.3	37.3	Max Throughput-50%Read	9.1	6,582.3	205.7	13.2
Random-8k-70%Read	171.5	301.2	2.4	27.6	Random-8k-70%Read	166.2	314.3	2.5	26.0

SAN Cache	1024MB	SCSI Controller	LSI		SAN Cache	1024MB
Block (Segment) size	256	VMFS block size	1MB		Block (Segment) size	256
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation	Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	2.5	20,206.6	631.5	84.6	Max Throughput-100%Read	3.1	19,257.7	601.8	18.7
RealLife-60%Rand-65%Read	162.3	338.7	2.6	22.5	RealLife-60%Rand-65%Read	159.4	354.2	2.8	19.1
Max Throughput-50%Read	8.8	6,627.7	207.1	38.1	Max Throughput-50%Read	9.1	6,573.3	205.4	13.8
Random-8k-70%Read	184.6	293.6	2.3	23.8	Random-8k-70%Read	176.6	310.4	2.4	21.5

Re: DAWN OF SSDs - it was about time... intel X25-E teaser :-)

typera — Wed, 25 Feb 2009 13:28:37 GMT

VM for my tests is Win2003 std server SP2 with current updates, 384MB RAM. For each block size test I created a 200Gb LUN and formatted it NTFS using default NTFS block size.

Win2003 install is vanilla as can be. I also reinstalled vmware tools cleanly just to make sure its all installed correctly.

Changing over to LSI didn't result any major changes:

SAN Cache	512MB	SCSI Controller	LSI
Block (Segment) size	512	VMFS block size	1MB
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	2.6	20,074.4	627.3	81.4
RealLife-60%Rand-65%Read	82.7	673.6	5.3	22.7
Max Throughput-50%Read	8.8	6,667.0	208.3	37.0
Random-8k-70%Read	94.3	571.9	4.5	26.1

SAN Cache	512MB	SCSI Controller	Buslogic
Block (Segment) size	512	VMFS block size	1MB
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	2.1	20,146.7	629.6	88.5
RealLife-60%Rand-65%Read	81.9	681.8	5.3	22.7
Max Throughput-50%Read	8.9	6,605.4	206.4	38.3
Random-8k-70%Read	94.1	574.0	4.5	26.2

Re: DAWN OF SSDs - it was about time... intel X25-E teaser :-)

Mnemonic — Wed, 25 Feb 2009 11:45:04 GMT

Hi, nice to see that kind of results..

Could you just publish the VM configuration in both environments? RAM, OS, Filesystem.

Also.. If you are using Windows Server 2003, it would only be fair to change the SCSI controller to LSI instead of BusLogic.

I Cant remember... Does Citrix use Parallel VIrtualized SCSI Drivers?

Re: Open inofficial storage performance thread

christianZ — Tue, 24 Feb 2009 08:49:27 GMT

Thanks for that input

Re: Open unofficial storage performance thread

christianZ — Tue, 24 Feb 2009 08:47:46 GMT

I've just noticed your test results on DS3400 - the numbers are very low - check your configurations!!

Re: DAWN OF SSDs - it was about time... intel X25-E teaser :-)

typera — Tue, 24 Feb 2009 06:47:08 GMT

Server: IBM x3650, 2 x Intel x5450 3Ghz QC; 28Gb RAM
SAN: IBM DS3200 (SAS) with 6 x 750GB SATA in RAID10 (1Gb Cache
however when upgraded from 512MB tests were not affected.. which was
good)
Tests: Five tests were carried out whilst varying the SAN's
logical drive segment (block) size from 32 through to 512 (max
supported on DS3200). No block alignment was done or any other
performance tuning with either environment.

VMWare ESX 3.5 Update 3

Citrix XenServer Enterprise 5.0.0

SAN Cache	512MB	SCSI Controller	Buslogic		SAN Cache	512MB
Block (Segment) size	512	VMFS block size	1MB		Block (Segment) size	512
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation	Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	2.1	20,146.7	629.6	88.5	Max Throughput-100%Read	4.3	14,202.2	443.8	20.4
RealLife-60%Rand-65%Read	81.9	681.8	5.3	22.7	RealLife-60%Rand-65%Read	81.3	691.8	5.4	19.7
Max Throughput-50%Read	8.9	6,605.4	206.4	38.3	Max Throughput-50%Read	8.5	6,999.9	218.7	13.9
Random-8k-70%Read	94.1	574.0	4.5	26.2	Random-8k-70%Read	92.5	588.7	4.6	23.1

SAN Cache	512MB	SCSI Controller	Buslogic		SAN Cache	512MB
Block (Segment) size	256	VMFS block size	1MB		Block (Segment) size	256
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation	Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	17.0	3,527.8	110.2	23.2	Max Throughput-100%Read	4.3	14,023.4	438.2	21.3
RealLife-60%Rand-65%Read	246.8	242.5	1.9	14.2	RealLife-60%Rand-65%Read	82.9	684.2	5.3	18.6
Max Throughput-50%Read	35.4	1,696.7	53.0	19.1	Max Throughput-50%Read	9.0	6,661.1	208.2	14.3
Random-8k-70%Read	331.5	180.7	1.4	14.1	Random-8k-70%Read	94.0	587.4	4.6	21.5

SAN Cache	512MB	SCSI Controller	Buslogic		SAN Cache	512MB
Block (Segment) size	128	VMFS block size	1MB		Block (Segment) size	128
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation	Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	17.4	3,443.4	107.6	24.2	Max Throughput-100%Read	3.9	15,616.5	488.0	20.8
RealLife-60%Rand-65%Read	253.9	236.1	1.8	14.1	RealLife-60%Rand-65%Read	85.9	660.0	5.2	18.2
Max Throughput-50%Read	35.6	842.2	52.5	19.7	Max Throughput-50%Read	8.8	6,802.0	212.6	14.2
Random-8k-70%Read	334.5	179.1	1.4	13.9	Random-8k-70%Read	93.5	579.6	4.5	24.5

SAN Cache	512MB	SCSI Controller	Buslogic		SAN Cache	512MB
Block (Segment) size	64	VMFS block size	1MB		Block (Segment) size	64
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation	Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	17.9	3,345.0	104.5	23.6	Max Throughput-100%Read	4.0	15,077.8	471.2	20.3
RealLife-60%Rand-65%Read	263.7	226.6	1.8	14.0	RealLife-60%Rand-65%Read	90.1	627.7	4.9	18.7
Max Throughput-50%Read	35.6	1,682.8	52.6	18.9	Max Throughput-50%Read	8.8	6,792.7	212.3	14.4
Random-8k-70%Read	339.2	176.0	1.4	13.8	Random-8k-70%Read	99.6	551.4	4.3	22.4

SAN Cache	512MB	SCSI Controller	Buslogic		SAN Cache	512MB
Block (Segment) size	32	VMFS block size	1MB		Block (Segment) size	32
Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation	Test name	Avg Resp Time ms	Av IOs/sec	Av MB/sec	CPU Utilisation
Max Throughput-100%Read	20.6	2,913.2	91.0	23.0	Max Throughput-100%Read	4.1	14,916.9	466.2	20.2
RealLife-60%Rand-65%Read	281.8	212.0	1.7	14.2	RealLife-60%Rand-65%Read	99.2	571.7	4.5	18.4
Max Throughput-50%Read	35.8	1,675.6	52.4	18.7	Max Throughput-50%Read	9.1	6,503.1	203.2	15.8
Random-8k-70%Read	353.7	169.6	1.3	13.3	Random-8k-70%Read	111.3	508.8	4.0	18.6

Rather then adding to this already long thread, I suggest discussion re this specific results please go to http://communities.vmware.com/thread/196031

Re: DAWN OF SSDs - it was about time... intel X25-E teaser :-)

Mnemonic — Mon, 23 Feb 2009 12:45:14 GMT

Ok .. Sounds like a plan.. Hope to see SSD's with more run time in the future..

Re: DAWN OF SSDs - it was about time... intel X25-E teaser :-)

radimf — Mon, 23 Feb 2009 12:24:09 GMT

Hi,
Thanks for info.

With todays technology - SUN has very interesting aproach - you could
use SSDs as L2ARC cache, which works as tier 0 for read operations.
So if you get a checksum error on read - ZFS reads for physical drives.
Which only incurs latency penalty - you do not loose any data.
If read errors start to mount.... Drive shuts its self down, so you are
back on mechanical drive IO level.

If I use Intel enterpreis SSDs - I plan to use only smaller part as
partition - that way drive gets more life, with planned replacement
within 2 years.

Re: DAWN OF SSDs - it was about time... intel X25-E teaser :-)

Mnemonic — Mon, 23 Feb 2009 11:59:28 GMT

Hi radimf,

I just spend some time today reading up on SSD's, and found this that might be of interest to you..

"Customers, such as a large OEM we won't mention, have been trying to validate flash SSDs for enterprise applications by looping hardcore I/O loads, and they all failed with write errors after only a few months." - Toms Hardware

http://www.tomshardware.com/reviews/ssd-hdd-flash,2127.html

The problem is that eventhough SSD's are more reliable because of the No-Moving-Parts fact, the sectors can only be changed roughly about 100.000 times, before they fail. (Current Drives) With IO intensive applications like Virtual Machines OS drives, this might cause problems.

I think that this might be a problem, when running ESX, especially it the SSD's are used in a SAN structure, where a controller might spend a lot of time rearranging the data, and thereby slowly "destroying" the disks.

Just something for you to consider.. Maybe SSD's are not ready for production ESX environments.

Re: Open unofficial storage performance thread

oreeh — Mon, 09 Feb 2009 07:19:05 GMT

Note to all posters in this thread:

Please don't use long lines consisting only of "#######" or "+++++++", as this will break the thread display.

Oliver Reeh
VMware Communities User Moderator

Re: Open inofficial storage performance thread

black33 — Thu, 29 Mar 2007 12:50:12 GMT

+++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS / PE2950 LOCAL STORAGE
+++++++++++++++++++++++++++++++++++++

SERVER TYPE: Windows 2003 Std SP1 - 512MB RAM
CPU TYPE / NUMBER: 1 vCPU
HOST TYPE: Dell PE2950, 16GB RAM, 2 x quadcore Xeon X5355
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Perc 5/i - 6 x 300GB SAS 15Krpm / 1MB VMFS3

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........____3.06____..........___13959___.........___438____

RealLife-60%Rand-65%Read......____27.8____..........___1711___.........____8.31____

Max Throughput-50%Read..........____5.71____..........___10343___.........____323____

Random-8k-70%Read.................____25.4____..........___1631___.........____6.70____

Re: Open inofficial storage performance thread

mittell — Thu, 29 Mar 2007 20:39:48 GMT

Borrowed an HP MSA1510i for a couple of weeks. First tests don't look too promising...this was with the SATA enclosure (MSA20) attached, also have an SCSI enclosure (MSA30) with a similar number of spindles (10k I believe) so will post how much improvement the better disks give.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: Windows 2003 R2 VM 1024MB RAM 10GB VMDK.
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell SC1430, 2GB RAM; 1 XEON 5120, 1,86 GHz, DualCore
STORAGE TYPE : MSA1510i Read Cache:128Mb Write Cache 128MB RAID5 7+1 SATA Disks (MSA20 cabinet). Single Controller.
CONNECTION: 2 x 1GBe, dedicated Cisco 3750 switch.

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........16.93511..........3473.355..........108.5424

RealLife-60%Rand-65%Read.....192.419944.........306.6018.......2.395327

Max Throughput-50%Read.......... 28.906212.......... 2054.455.......64.20172

Random-8k-70%Read................. 196.450229.........295.3122.......2.307127

CPU Util.-XX%; 65.449696% - 15.945951% - 30.388721% - 17.881071%

4GB IOmeter Datafile

Re: Open inofficial storage performance thread

CWedge@Amsa — Mon, 23 Apr 2007 12:11:41 GMT

Yes indeed it seems to be the fastest for now (and
everlasting ?).

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

IBM DS8300:
RealLife-60%Rand-65%Read......___1.42__..........__260
3__.........__31.30___

EMC DMX3000:
RealLife-60%Rand-65%Read......___1.8_____..........___
2170___.........___17_____

The question here could be which system could serve
more guests with such performance .

They both are playing of course in other league - can
profit from the large cache (32 GB).

I notice the MB/s number of 31.10 - that seems to be
a bug (by block size 8kB).

I was running 180 VM's over 9TB's while that test was running....
Up till a little while ago I was getting errors "Low I/O Detected" lololol....

I love this thing...No matter how hard I push the Vm's it's still falling asleep.
Which I think makes the ESX hosts life a lot easier as they don't have to buffer so much I/O's which keeps CPU down.

I just keeping running into low memory errors, never any cpu or i/o errors.

One other side note. I only have 144 of the 572 Drives it can Hold ;-p
Although i think it'd be neck and neck running these two boxes at full capacity.

Also keep in mind now our DMX3000 is only DMX-2 Archietecture...
The DMX3 is even faster and more scalable...Hopefully I'll have one of those this year...

Message was edited by:
CWedge@Amsa

Re: Open inofficial storage performance thread

christianZ — Fri, 12 Sep 2008 17:56:44 GMT

+++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS / PRODUCTION INFRASTRUCTURE
+++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM ON ESX 3.5 Upd. 1
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell PE2950, 16GB RAM; 2x XEON X5450, 3,0 GHz, QC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 1 / 14+2 SAS10k / R50 (FW 3.3.1 w/o thinprov.)
SAN TYPE / HBAs : iSCSI, QLA4050 HBA

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........__17______..........___3540___.........___110____

RealLife-60%Rand-65%Read......___36_____..........___1590___.........____12____

Max Throughput-50%Read..........____11____..........___5620___.........___175____

Random-8k-70%Read.................____39____..........___1468___.........____12____

EXCEPTIONS: VCPU Util. <22%, all tests running on prod. systems; storage unit (1) cap. utilization > 50 %;
runnings machines: vms ca. 45/(ca. 100 configured), ph. ca. 5, in production since 03.2007

These tests are running on not stripped volume (only one member) - it is a large difference
to the first test on fresh system. I wonder.

First what I noticed here was if I make the test on a new (free) esx host w/o any vm running on it then
I can reach much more (~2300 ios/sec by Reallife) - i.e. the performance degradation comes with
more running vms (on esx), not with storage saturation.

And I see a lot of new entries here from:
dalepa - interesting nfs analysis
saqp - test of Xtravirt xvs virtual san
happyhammer (again) - test of cx3-40
chrisfmss - test of Eql PS5000 with satas

Thanks to all for keeping this thread alive.

Re: Open inofficial storage performance thread

cmanucy — Tue, 30 Oct 2007 21:24:19 GMT

Part 3: Running 8 tests at one time on 2 ESX hosts (each with 4 VM's, each clones of each other).
The back-end for this test is split - I used 3 LeftHand DL320s, 36 x 15K 300GB SAS drives at
RAID 50 (at the host level) and then a both one-way (no replication) and two-way replicated volumes
for EACH host. (Eg, each host has its own volume to run off of). All tests run with a 3COM 4500G
as the switch.

All tests run on a non-formatted volume using Microsoft iSCSI initaitor.

Sessions 1 -> 4 were on a 1-Way volume.

Session 1:
Frontend: IBM x3500, 2 Xeon 5140's, 5GB RAM, 2xIntel PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	32.990719	1,799.65	56.239195	52.94
RealLife-60%Rand-65%Read	44.732703	914.94	7.147985	58.46
Max Throughput-50%Read	29.608262	1,961.87	61.308413	54.70
Random-8k-70%Read.	44.877465	1,026.99	8.023367	50.79

Session 2:
Frontend: IBM x3500, 2 Xeon 5140's, 5GB RAM, 2xIntel PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	32.855391	1,806.96	56.467508	51.66
RealLife-60%Rand-65%Read	45.451255	909.93	7.108827	57.28
Max Throughput-50%Read	31.814684	1,865.16	58.286388	49.85
Random-8k-70%Read.	45.038477	1,015.53	7.933811	51.18

Session 3:
Frontend: HP DL385, 12GB RAM, 2xOpteron 2.6 DC, 2xIntel e1000 PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	38.523787	1,502.81	46.962704	50.29
RealLife-60%Rand-65%Read	46.062629	881.17	6.884125	57.69
Max Throughput-50%Read	62.866689	968.49	30.265301	31.65
Random-8k-70%Read.	45.212364	1,002.04	7.828432	51.38

Session 4:
Frontend: HP DL385, 12GB RAM, 2xOpteron 2.6 DC, 2xIntel e1000 PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	55.957563	1,070.46	33.451978	41.64
RealLife-60%Rand-65%Read	56.890474	790.74	6.177638	50.33
Max Throughput-50%Read	107.05434	554.62	17.331824	26.53
Random-8k-70%Read.	57.601321	830.74	6.490151	46.29

Sessios 5 -> 8 were on a 2-way volume:

Session 5:
Frontend: IBM x3500, 2 Xeon 5140's, 5GB RAM, 2xIntel PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	47.009389	1,268.83	39.650974	48.48
RealLife-60%Rand-65%Read	49.853864	878.28	6.861588	53.10
Max Throughput-50%Read	87.856369	691.62	21.613142	28.12
Random-8k-70%Read.	56.399108	834.95	6.523053	48.27

Session 6:
Frontend: IBM x3500, 2 Xeon 5140's, 5GB RAM, 2xIntel PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	47.319051	1,261.90	39.43428	49.67
RealLife-60%Rand-65%Read	104.975759	446.25	3.48632	42.61
Max Throughput-50%Read	54.936891	1,069.91	33.434719	43.62
Random-8k-70%Read.	109.253424	419.66	3.278631	44.73

Session 7:
Frontend: IBM x3500, 2 Xeon 5140's, 5GB RAM, 2xIntel PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	27.050811	2,209.13	69.03526	53.81
RealLife-60%Rand-65%Read	87.214384	508.69	3.974111	46.15
Max Throughput-50%Read	37.742076	1,578.58	49.330604	42.86
Random-8k-70%Read.	80.979612	535.10	4.180488	48.50

Session 8:
Frontend: IBM x3500, 2 Xeon 5140's, 5GB RAM, 2xIntel PCI-X teamed NICs, VMWare ESX 3.0.2, Win2K3 SP2 VM, 1MB BS, MS iSCSI Initiator

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	38.51754	1,497.39	46.793515	49.15
RealLife-60%Rand-65%Read	113.306869	416.15	3.25118	41.35
Max Throughput-50%Read	43.552626	1,301.15	40.660856	46.37
Random-8k-70%Read.	119.105131	396.74	3.099504	40.83

Result summary:

Session Totals From 1-Way Sessions (1 ->4) - Response Time & CPU Use is AVERAGE, I/O's and MB/sec are SUMS

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	40.081865	6,179.88	193.12139	49.13
RealLife-60%Rand-65%Read	48.28426525	3,496.78	27.318575	55.94
Max Throughput-50%Read	57.83599375	5,350.14	167.19193	40.68
Random-8k-70%Read.	48.18240675	3,875.30	30.275761	49.91

Session Totals From 2-Way Sessions (5 ->8) - Response Time & CPU Use is AVERAGE, I/O's and MB/sec are SUMS

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	39.97419775	6,237.25	194.91403	50.44
RealLife-60%Rand-65%Read	88.44531931	4,867.86	38.030186	46.51
Max Throughput-50%Read	48.51689669	9,299.78	290.61811	43.38
Random-8k-70%Read.	89.38014344	5,226.80	40.834384	45.99

Session Totals From All Sessions (1 ->8) - Response Time & CPU Use is AVERAGE, I/O's and MB/sec are SUMS

TEST NAME	Av. Resp. Time ms	Av. IOs/sek	Av. MB/sek	CPU Use
Max Throughput-100%Read.	40.02803138	12,417.13	388.03541	49.79
RealLife-60%Rand-65%Read	68.36479228	8,364.64	65.348761	51.22
Max Throughput-50%Read	53.17644522	14,649.92	457.81003	42.03
Random-8k-70%Read.	68.78127509	9,102.10	71.110145	47.95

This was the best 'speed' test of all the bunches. With 388+ MB/sec for the 100% test, the MS iSCSI initator bested the ESX initator by about 60MB/sec with a similar test run. Also, with 12,400+ I/O sec - that's not too shabby. Best ESX iSCSI initator could pull with an 8-session test was just under 10,000. Random test was almost 20MB/sec faster as well - using ESX initiator, best result from an 8-session test was 45 MB/sec (with almost 6,000 I/O sec).

Re: Open inofficial storage performance thread

chrisfmss — Mon, 15 Sep 2008 15:45:53 GMT

I have made another test with 4 nic and 2 ps5000e. It doesn't make such diffirence.

SERVER TYPE: PHYS.
CPU TYPE / NUMBER: XEON 3.2GHz / 1
HOST TYPE: Proliant ML350 G3, 4GB RAM; 1x XEON, 3.2 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS5000E x 2 / 2 x (14+2) 1TB Disks SATA / R10
MS Initiator, flow control enable and jumbo frame enable. 4 x 1GB Nics

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........___8.4___..........__6180.46__.........___193.14___

RealLife-60%Rand-65%Read......____4.6___.........._11118.87__.........___86.87___

Max Throughput-50%Read..........___8.3____..........__6288.04__.........__196.50__

Random-8k-70%Read.................___4.3____.........._12044.71__.........__94.10___

EXCEPTIONS: CPU Util.-XX%;

Re: Open inofficial storage performance thread

asp24 — Sun, 21 Sep 2008 10:20:24 GMT

SERVER TYPE: W2003 VM on ESX3.5.0.
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Tyan Transport TA26, 32GB RAM; 2x quad core AMD 2352, 2,1 GHz
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Open-E DSS,
u160 external sata-scsi Infortrend A16U-G1A3 / 12+1 Disks / R5
VMFS 1900GB LUN, 1MB block size, SW Iscsi init.
Samsung F1 750GB disks, Dual Opteron 2 GHz, 4GB mem, LSI U320 controller on Open-E server.

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........___17.23.........._3334.96__........._104.21___

RealLife-60%Rand-65%Read......___76.39__..........__747.13__.........__5.84___

Max Throughput-50%Read..........__13.33__..........__4087.47__........._127.73__

Random-8k-70%Read.................___89.29__.........._643.06___.........__5.02__

This will soon be used for DR storage.

I have ordered an INFORTREND 3U 16-Bay iSCSI-to-SAS RAID ( S16E-G1430 ). I have also ordered
9x Seagate Cheetah 15k.5 300GB SAS drives and 9x Samsung F1 1TB drives (one spare of each type)

I hope I will be satisfied with the new Infortrend box. 4 x Gigabit's for ISCSI traffic. I think I will use
2x dedicated gigabit NICs on each host for iscsi traffic (6 hosts total, about 60 VM's,
mix of web, sql, exchange and terminal servers (RDP) )

I have not decided on RAID level for the 8 SAS-drives yet. Raid 5, 10 or 50. Comments?

Message was edited by: oreeh

Reason: fixed formatting to prevent thread display issues

Oliver Reeh
VMware Communities User Moderator

Re: Open inofficial storage performance thread

jasonboche — Mon, 09 Feb 2009 06:57:38 GMT

+++++++++++++++++++++++++++
TABLE OF RESULTS
+++++++++++++++++++++++++++

SERVER TYPE: Windows XP VM ON ESXi 3.5.0u3
CPU TYPE / NUMBER: vCPU / 1
HOST TYPE: HP Proliant DL385G2, 16GB RAM; 2x DC AMD OPTERON 2218 2.6GHZ (4 CORES TOTAL)
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Openfiler 2.2 swiSCSI on HP Proliant ML570G2,
4x P4 2.xGHz 9.5GB RAM, 12x136GB 10kRPM RAID5
SAN TYPE / HBAs : n/a

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........__18______..........___3238___.........___101____

RealLife-60%Rand-65%Read......___135_____..........___434___.........____3____

Max Throughput-50%Read..........____91____..........___651___.........___20____

Random-8k-70%Read.................____143____..........___408___.........____3____

EXCEPTIONS: VCPU Util. 33, 17, 17, 18 %;

SERVER TYPE: Windows XP VM ON ESXi 3.5.0u3
CPU TYPE / NUMBER: vCPU / 1
HOST TYPE: HP Proliant DL385G2, 16GB RAM; 2x DC AMD OPTERON 2218 2.6GHZ (4 CORES TOTAL)
STORAGE TYPE / DISK NUMBER / RAID LEVEL: fibre channel SAN, 7x36GB 10kRPM RAID5
SAN TYPE / HBAs : MSA1000 fibre channel SAN, Active/Active controller firmware 7.00, QLA2432 HBA

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........__12______..........___4864___.........___152____

RealLife-60%Rand-65%Read......___110_____..........___447___.........____4____

Max Throughput-50%Read..........____16____..........___3737___.........___117____

Random-8k-70%Read.................____87____..........___620___.........____5____

EXCEPTIONS: VCPU Util. 38, 34, 32, 27 %;

SERVER TYPE: Windows XP VM ON ESXi 3.5.0u3
CPU TYPE / NUMBER: vCPU / 1
HOST TYPE: HP Proliant DL385G2, 16GB RAM; 2x DC AMD OPTERON 2218 2.6GHZ (4 CORES TOTAL)
STORAGE TYPE / DISK NUMBER / RAID LEVEL: fibre channel SAN, 7x36GB 15kRPM RAID5
SAN TYPE / HBAs : MSA1000 fibre channel SAN, Active/Active controller firmware 7.00, QLA2432 HBA

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........__12______..........___5019___.........___157____

RealLife-60%Rand-65%Read......___71_____..........___711___.........____6____

Max Throughput-50%Read..........____19____..........___3108___.........___97____

Random-8k-70%Read.................____56____..........___961___.........____8____

EXCEPTIONS: VCPU Util. 39, 33, 29, 28 %;

Jason Boche
boche.net - VMware Virtualization Evangelist
VMware Communities User Moderator
Minneapolis Area VMware User Group Leader

Re: Open unofficial storage performance thread

stirgin — Sat, 13 Dec 2008 16:35:16 GMT

[SERVER TYPE: Host with ESX 3.5

CPU TYPE / NUMBER: Dual 2.0 GHz Quad Core

HOST TYPE: Dell 2950

STORAGE TYPE / DISK NUMBER / RAID LEVEL: Local storage 4 x 750GB SATA

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sec-------Av. MB/sec------

Max Throughput-100%Read........____0.389__.........._____5473_.........___171__

RealLife-60%Rand-65%Read......._____109__.........._____327.46__.........____2.55__

Max Throughput-50%Read.........______.701__..........____4956__.........____154.89__

Random-8k-70%Read.............._____122.39__.........._____255.67__........._____1.99__

EXCEPTIONS:

SERVER TYPE: Host with ESX 3.5

CPU TYPE / NUMBER: Dual 2.0 GHz Quad Core

HOST TYPE: Dell 2950

STORAGE TYPE / DISK NUMBER / RAID LEVEL: Local storage 6 x 750GB SATA

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sec-------Av. MB/sec------

Max Throughput-100%Read........____0.376__.........._____5756_.........___179__

RealLife-60%Rand-65%Read......._____88__.........._____456__.........____3.56__

Max Throughput-50%Read.........______.559__..........____5410__.........____169__

Random-8k-70%Read.............._____96.24__.........._____357__........._____2.78__

EXCEPTIONS:

SERVER TYPE: Host with ESX 3.5

CPU TYPE / NUMBER: Dual 1.6 GHz Quad Core

HOST TYPE: Dell 2950

STORAGE TYPE / DISK NUMBER / RAID LEVEL: Local storage 6 x 300GB SAS

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sec-------Av. MB/sec------

Max Throughput-100%Read........____4.6__.........._____1355_.........___42.35__

RealLife-60%Rand-65%Read......._____23.18__.........._____500.45__.........____3.9__

Max Throughput-50%Read.........______5.11__..........____1258.88__.........____39.34__

Random-8k-70%Read.............._____21.43__.........._____496.61__.........___3.879____

EXCEPTIONS:

SERVER TYPE: Host with ESX 3.5

CPU TYPE / NUMBER: Dual 2.0 GHz Quad Core

HOST TYPE: Dell 2950

STORAGE TYPE / DISK NUMBER / RAID LEVEL: NAS Connected (1GB copper), not isolated lan - NetAPP 3050 20x300GB SAS

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sec-------Av. MB/sec------

Max Throughput-100%Read........____.42__.........._____4862_.........___151.94__

RealLife-60%Rand-65%Read.......___20.50___.........._____501.79__.........____3.92__

Max Throughput-50%Read.........______.58__..........____4469__.........____139.67__

Random-8k-70%Read.............._____24.79__.........._____501.705__........._____3.91__

One thing I noticed is that I can't seem to get above ~500 IO's in the random tests,
I used the config file posted early in this thread, and all the setting look right...
maybe I'm missing something?

Message was edited by: oreeh

Reason: fixed the formatting to prevent thread display issues

Oliver Reeh
VMware Communities User Moderator

Re: DAWN OF SSDs - it was about time... intel X25-E teaser :-)

radimf — Sun, 01 Feb 2009 23:34:28 GMT

Hi to all.

I am newbie around here. Sorry for the provocative subject I hope I interpreted results right ...
I am solving storage / architecture dilema, which I will be fully posting in another thread on this
enterpreis forum.

To get a picture why I did this test.

I am looking for solid and fast solution for our 14 VMs consolidation (any array should be ok) and
possibly for our 1 production SQL 2005 database (planned as 4 vCPUs, 32GB RAM, 120 GB DB
for ERP and controlling)

Question is - to go with traditional FC SAN (*HP EVA 4400*, 16 FC 15krpm disks in 2 trays - my
maximum budget) , or leave those FC mamoths behind, and go with *Lefthand, or Falconstor virtual
appliances on 2 ESX servers* (those will host SQL, and other VMs) *with 10gbit crossover replication
for data HA*? So I could move 1 server to nearby DR site.

One more possibility is to use those 2 ESXs with virtual storage appliances as "iSCSI SAN" and
to have main production SQL server physical 2xQC 32GB RAM.

4 vs. 8 CPUs and no IO overhead if we use FC SAN, or with iSCSI overhead.

ANY comments welcommed.

*HP EVA 4400 16x FC 15krpm 146GB vs. 10Gbit clustered 2x ESX virtual appliance from LEFTHAND,
or FALCONSTOR with possible SSD RAID AS TIER 0*

Can I assume, that 4x raid5 (yes raid 5 works superb with SSDs even for writes) Intel x25-E will smoke
my max. EVA possible config (for only storage critical 4cpu SQL VM)?

Sorry for long start. NOW THE RESULTS for 1 INTEL X25-E 32GB SSD - MAX. THROUGHPUT is not
interesting it is just one SATA disk - random numbers are :-))))

+++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
+++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: Win XP 64bit VM on ESXi 3.5

CPU TYPE : intel QC 6600

/ NUMBER: VCPU / 2 (6 other low IO, low CPU VMs running on this ESXi 3.5)

HOST TYPE: Asus WS PRO x38 ICH9-R, 8GB RAM; 1x intel 6600, 2,40 GHz, QC
STORAGETYPE/ DISK NUMBER / RAID LEVEL: none/ 1x Intel X25-E SSD 32GB

VMFS 32 GB LUN , 1 MB

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........_1.38_..........__3173.78__.........___99.18__ __CPU *23.92*%!!!

RealLife-60%Rand-65%Read......__1.87__..........___2937.336__.........__22.94___CPU 16.88%

Max Throughput-50%Read.........._4.21_..........__1939.38__.........____60.60____CPU 15.00%

Random-8k-70%Read.................____1.62______.........._3024.72........._23.63____CPU 17.68%

EXCEPTIONS: CPU Util.-23,17,15,18%!!!;

I will try to setup next week trial of Falconstor vmware appliance using Intel SSD, and test it from our
old production SQL SERVER via iSCSI.

Message was edited by: oreeh

Reason: fixed the formatting to prevent thread display issues

Oliver Reeh
VMware Communities User Moderator

Re: OPEN INOFFICIAL STORAGE PERFORMANCE THREAD

mitchellm3 — Tue, 11 Mar 2008 14:29:20 GMT

SERVER TYPE: Win2003 Std. VM ON ESX 3.0.2 update1
CPU TYPE / NUMBER: VCPU / 2
HOST TYPE: HP DL585 G2, 32GB RAM, 4x AMD 8222 SE, 3.0 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: IBM SVC 2145-8G4 in front of
DS4800 (7)shelves 4Gbit/sec 146GB drives. (1) 450GB VMFS / striped
SAN TYPE / HBAs : 4G FC Qlogic

*TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------ *

Max Throughput-100%Read........____4.79____.........._11257___.........__351____

RealLife-60%Rand-65%Read......____2.47____.........._8457___.........___66____

Max Throughput-50%Read..........___0.89____.........._12199___.........___381____

Random-8k-70%Read.................___0.43____.........._16359___.........___127____

EXCEPTIONS: CPU 45%, 46%, 51%, 50%. without second proc, cpu is 99%

Re: OPEN INOFFICIAL STORAGE PERFORMANCE THREAD

christianZ — Sat, 03 Mar 2007 19:18:04 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS PHYS. WIN2003 / MS ISCSI INITIATOR / 2 X PS3600X
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: Win2003 Std. PHYS.
CPU TYPE / NUMBER: CPU / 2
HOST TYPE: Dell PE1950, 4GB RAM, 2x XEON 5130, 2,0 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 2 / 2x14+2x2 SAS10k / R50 / LUN STRIPPED
SAN TYPE / HBAs : iSCSI; iSCSI softw. initiator in WIN2003, 1XGb NIC for iSCSI

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........____17____..........___3570___.........___112____

RealLife-60%Rand-65%Read......_____8____..........___5919___.........____46____

Max Throughput-50%Read..........____12____..........___4810___.........____150___

Random-8k-70%Read................._____8____..........___5720___.........____45____

EXCEPTIONS: Seq. r/w not as high as by single box

Re: OPEN INOFFICIAL STORAGE PERFORMANCE THREAD

christianZ — Sat, 03 Mar 2007 15:20:47 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS PHYS. WIN2003 / MS ISCSI INITIATOR
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: Win2003 Std. PHYS.
CPU TYPE / NUMBER: CPU / 2
HOST TYPE: Dell PE1950, 4GB RAM, 2x XEON 5130, 2,0 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 1 / 14+2 SAS10k / R50
SAN TYPE / HBAs : iSCSI; iSCSI softw. initiator in WIN2003, 1XGb NIC for iSCSI

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........____17____..........___3560___.........___112____

RealLife-60%Rand-65%Read......____13____..........___3550___.........____28____

Max Throughput-50%Read..........____10____..........___5950___.........____186___

Random-8k-70%Read.................____14____..........___3340___.........____26____

EXCEPTIONS:

Re: OPEN INOFFICIAL STORAGE PERFORMANCE THREAD

GabVanZ — Sun, 08 Feb 2009 23:22:57 GMT

SERVER TYPE: VM ON ESX 3.5u3

CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Whitebox 1x Q6600 (quad core), 8GB RAM; 1x Q6600 2,4 GHz

STORAGE TYPE / DISK NUMBER / RAID LEVEL: local SATA-2 disk / 1 disk / no raid

SAN TYPE / HBAs : local storage

TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------

Max Throughput-100%Read........__16.79______..........___3544___.........___110.78____

RealLife-60%Rand-65%Read......___368_____..........___158___.........____1.24____

Max Throughput-50%Read..........____255____..........___231___.........___7.29____

Random-8k-70%Read.................____497____..........___117___.........____0.92____

Re: Open inofficial storage performance thread

radimf — Mon, 09 Feb 2009 07:09:14 GMT

Hi,

please could you test your MSA 1000 FC and Openfiller iSCSI in a raid 10 config?

IF it was possible. Thanks

I uploaded results from one Intel SSDs earlier (still early adopter technology), but have a look at my 2nd test - it is simply another league to mechanical HDDs (response time, random numbers).

Low capacity, but it rocks.

I am in the middle of testing Falconstor virtual appliance (it has superb replication capabilities) - it is not that far from raw device numbers (as soon as I battle through NIC iSCSI teaming - it will rock

That way we could get enterpreis features virtual HA SAN (with SQL consistent snapshots), on 2 ESX hosts with RAIDed Intel SSDs.

Re: Open unofficial storage performance thread

darren.wardle50 — Sun, 28 Sep 2008 23:08:48 GMT

I cannot locate the values for Av. IOs/sek and Av. MB/sek are in the csv file or in IO Meter
If someone could point them out to me in the attached file I will update the results below.

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........4.16.....................................______

RealLife-60%Rand-65%Read......71.98...................................________

Max Throughput-50%Read........13.38....................................________

Random-8k-70%Read.............66.73...................................._______

EXCEPTIONS: CPU Util. 57%-42%-25%-24%;

##################################################################################

Message was edited by: oreeh

Reason: fixed formatting to prevent thread display issues

Oliver Reeh
VMware Communities User Moderator

Re: INTEL X25-E 32gb RETESTED - PHYSICAL DRIVE

Mnemonic — Wed, 04 Feb 2009 10:07:59 GMT

Try to reboot VM after each test. Simetimes I have seen wird and incosistent results if I run the tests right after each other.

Re: INTEL X25-E 32gb RETESTED - PHYSICAL DRIVE

radimf — Mon, 02 Feb 2009 22:04:18 GMT

Hi,

my initial numbers from INTEL X25-E 32gb I plan to use in RAID config for Falconstor or Lefthand were nice, but something looked strange (max. throughput numbers were to low)

Here are NEW numbers from raw device - VMFS 1mb , I deleted partition, so no NTFS this time. IOMETER run against PHYSICAL DRIVE

(may be it was due to SSD slowing after high performance demand - as observed first on forum.ssdworld.ch)

olso interesting -Av. Resp. Time ms got way higher - any comments welcome

CPU TYPE : intel QC 6600

/ NUMBER: VCPU / 2 (6 other low IO, low CPU VMs running on this ESXi 3.5)

HOST TYPE: Asus WS PRO x38 ICH9-R, 8GB RAM; 1x intel 6600, 2,40 GHz, QC
STORAGETYPE/ DISK NUMBER / RAID LEVEL: none/ 1x Intel X25-E SSD 32GB NO PARTITION

VMFS 32 GB LUN , 1 MB
##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........_11.32_..........__5183.93__........._161.99__ __CPU *20.25*%!!!

RealLife-60%Rand-65%Read......__16.02__.........._3646.58__.........__28.48___CPU 19.51%

Max Throughput-50%Read.........._30.67_.........._1939.25__.........____60.60____CPU 12.47%

Random-8k-70%Read.................____14.11______.........._4167.31........._32.55____CPU 17.97%

EXCEPTIONS: CPU Util.-20,19,12,18%!!!

Re: Open unofficial storage performance thread

radimf — Mon, 02 Feb 2009 12:31:17 GMT

Hi,

thanks for response. I am realy interested what people have to say about this .

I will retest this drive with partiton aligment and different NTFS cluster sizes, may be it will bring more interesting numbers.

After that I will dig into Falconstor virtual appliance performance SSD as tier0 and rest with classic mechanical drives. Lefthand is my another choice.

If I find no obstacles ahead - I plan to continue to clustered config via 1gbit iSCSI for data redundancy.

If it works - HP people will have a tough time to sell me on HP EVA4400 for our install.

Except you guys with SAN expertise bring me down to earth, and influence me to go for safer EVA or CX solution

If we needed 25TB storage space - no contest - traditional SAN rules, but for our max. 3TB of data and only 120GB of performace critical data ....

Re: Open unofficial storage performance thread

Mnemonic — Mon, 02 Feb 2009 11:53:15 GMT

This is really a reply to radimf, but for some reason I cannot reply to his post.

It is certaintly a interesting senario. It is hard to say anything intelligent about this, other then those performance number from the SSD disk are kicking SAS in the behind! If those SSD disks can really pull of those numbers, and the disk space are sufficiant, it is worth a test.

Remember to compare ease-of-use when a drive fails, or when replication between the ESX Virtual Storage Appliance servers dies and all recovery senarios that you can think of.

/Brian

Re: Open unofficial storage performance thread

eliot — Wed, 21 Jan 2009 09:30:46 GMT

I've attached a 1Mb ISO containing the IOMeter tool and Christian's Template.

Attach it to your VM and it will autorun and load up the template for you.

Re: Open unofficial storage performance thread

Remko N — Mon, 23 Jun 2008 15:28:08 GMT

Re: Open unofficial storage performance thread

christianZ — Sat, 17 Jan 2009 10:27:50 GMT

Such high latencies occuring again are not normal - your storage could be overloaded or you have some kind of problems (irq sharing, scsi reservations, incorrect san configuration, ...)

How many vms are configured on the same lun?

The best thing should be to start your own thread with your problem description.

Paste the output from "excfg-mpath -l", /var/log/mvkernel (with warnings/errors), decrtibe what storage you are using (how many disks, how configured), etc.

Re: Open unofficial storage performance thread

aworkman — Fri, 16 Jan 2009 17:06:36 GMT

What kind of max latency's are people seeing??

I have latency spikes of up to 3-5 seconds 3000-5000ms. Is this something that should be happening? I'm trying to isolate an array/fabric/vmware issue where we are getting reservation conflicts/failed io and so far high latency seems to be directly related to the time where the issues occur.

Re: Open unofficial storage performance thread

vmmarco — Wed, 07 Jan 2009 11:01:58 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM
VCPU / 1
HOST TYPE: HP DL360G5 10GB RAM; 2x XEON E5405 @ 2,00 GHz quad core
IBM DS3400 FC 4GB / 12+1 Disks / R5

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........____18______..........____3153______........._____98,5_____

RealLife-60%Rand-65%Read......____463______.........._____129_____.........____1______

Max Throughput-50%Read..........____407______.........._____144_____.........____4,5______

Random-8k-70%Read.................____449______.........._____133_____.........____1______

##################################################################################

Re: Open unofficial storage performance thread

Mnemonic — Mon, 12 Jan 2009 13:50:56 GMT

It looks like excel multiply your results with "1.000.000" Fix it or use openoffice.

Re: Open unofficial storage performance thread

Mnemonic — Mon, 12 Jan 2009 13:48:23 GMT

Your random number looks right. The reason for the low numbers in this test is that it does not test throughput but the performance of you disks, and since you have an array with a low number of disks you get poor performance. It is normal.

The same apply's to your results stirgin

Re: Open unofficial storage performance thread

christianZ — Sat, 10 Jan 2009 18:51:39 GMT

Copy and paste - simply copy a nice table (e.g. one of mine ), paste it (in "Normaltext" and insert your numbers overwriting the existing one. Check the look by "Preview". Done.

Re: Open unofficial storage performance thread

christianZ — Sat, 10 Jan 2009 18:35:35 GMT

For you definitely - as I mentioned in the past - for such monster storages (like yours) the best way should be to run the test on many vms simultaneously.

And don't forget to verify your results by "esxtop" (see my last postings).

Anyway thanks for joining in.

Reg

Christian

Re: Open unofficial storage performance thread

bertcord — Wed, 07 Jan 2009 23:11:58 GMT

In my opinion the default test file size is WAY too small. I ran the below tests on a volume that was 1,024GB

SERVER TYPE: Physical
CPU TYPE / NUMBER: Xeon X5355 2.66 Quad Core / 2
HOST TYPE: Dell 2950, 16GB RAM;
SAN Type: 3Par 4 Node T400 / Disks: 4GB FC 147GB 15k / RAID LEVEL: Raid10 / 240 Disks / Qlogic 2462 4Gbit FC HBA

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........___2.95_______..........___18,457_______.........___577_______

RealLife-60%Rand-65%Read......____4.52______..........____9,432______........._____73_____

Max Throughput-50%Read.........___1.82_______..........___28,259_______.........____883______

Random-8k-70%Read..............___3.8_______..........___10,772_______.........___81.82_______

##################################################################################

Re: Open unofficial storage performance thread

thor918 — Fri, 02 Jan 2009 19:53:42 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM ON ESXi 3.5.0 build 120505
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: IBM x346, 9GB RAM; 1x XEON , 3.6 GHz,
STORAGE TYPE / DISK NUMBER / RAID LEVEL: SCSI U320 3x146GB, raid5
SAN TYPE / HBAs : iSCSI, serveraid7k hardware raid controller

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........__13.1______..........___4502___.........___140.7____

RealLife-60%Rand-65%Read......___56.2_____..........____312___.........____2.4____

Max Throughput-50%Read..........__15.5____..........____3760___.........___117.5____

Random-8k-70%Read.................__57.7____..........___279.2___.........____2.2____

EXCEPTIONS:

Re: Open unofficial storage performance thread

liqu — Tue, 09 Dec 2008 11:02:54 GMT

Very poor random numbers Would you have any ideas on the cause of it?

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM on ESX 3.5.0 Update 3
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Sun Fire X4150, 16GB RAM; 2x XEON E5440, 2,83 GHz, QC; 10GbE
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Sun Fire X4250 8GB; 2xE5440; 10GbE / 10 Disks / R0 (Openfiler 2.3 over NFS (sync/wdelay))

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........__2.8__..........___11678___.........___366___

RealLife-60%Rand-65%Read......__71__..........___837___........._____6.5_____

Max Throughput-50%Read..........__1.1__..........___12011___.........___373____

Random-8k-70%Read.................___77___..........__774____........._____6_____

EXCEPTIONS: CPU Util.- 47%; 9%; 50%; 9%

##################################################################################

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM on ESX 3.5.0 Update 3
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Sun Fire X4150, 16GB RAM; 2x XEON E5440, 2,83 GHz, QC; 10GbE
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Sun Fire X4250 8GB; 2xE5440; 10GbE / 10 Disks / R0 (Openfiler 2.3 over iET iSCSI (write-back/blockio))

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........__5.4__..........___10273___.........___325___

RealLife-60%Rand-65%Read......__58__..........___1047___........._____8_____

Max Throughput-50%Read..........__8.2__..........___7325___.........___230___

Random-8k-70%Read.................___54___..........__1114__........._____10_____

EXCEPTIONS: CPU Util.- 30%; 9%; 50%; 9%

##################################################################################

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM on ESX 3.5.0 Update 3
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Sun Fire X4150, 16GB RAM; 2x XEON E5440, 2,83 GHz, QC; 10GbE
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Sun Fire X4250 8GB; 2xE5440; 10GbE / 10 Disks / R0 (Openfiler 2.3 over iET iSCSI (write-thru/blockio))

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........__5.4__..........___10023___.........___318___

RealLife-60%Rand-65%Read......__58__..........___1042___........._____8_____

Max Throughput-50%Read..........__6.1__..........___9530___.........___300___

Random-8k-70%Read.................___54___..........__1094__........._____9_____

EXCEPTIONS: CPU Util.- 30%; 14%; 25%; 17%

##################################################################################

Re: Open inofficial storage performance thread

christianZ — Thu, 20 Nov 2008 09:04:30 GMT

Check this thread here: http://communities.vmware.com/message/1103516#1103516

I suppose the IFT box works similar like MD3000i - each eth. port has its own ip address - so the way should be the same. Don't forget to input the right string (from your IFT type box) in your Esx (see attachment) - your string can vary - check with "esxcfg-info | grep -2 IFT", search for model don't forget the <blank>: at the end of the string.

Re: Open inofficial storage performance thread

asp24 — Thu, 20 Nov 2008 08:36:19 GMT

I have confirmed my numbers using ESXTOP. I get ca 110 MB/s most of the time, but there are some periods with MUCH less also.. The random IO tests also show between 4000 and 4500 IO's.

The VMs feel MUCH faster than with my old array (and the IO scores are 4-5x better), so I'm very happy with the Infortrend box.

I'm struggeling with multipathing, but I guess a support ticket is the right way to go there. But as I said earlier, the backup at night now runs at 100+ MB/s from two hosts at the same time (connected to 2 different vlans and nics on the Infortrend) giving me a total of 200 MB/s + backup at night, and the VM's are still fast while running backup.

If any of you are running Infortrend iscsi with multipathing, please share your settings!

Re: Open unofficial storage performance thread

christianZ — Thu, 20 Nov 2008 08:09:11 GMT

You can see that all in the results window (in your iometer screen).

Re: Open inofficial storage performance thread

christianZ — Thu, 20 Nov 2008 08:03:28 GMT

Fot those who didn't know - on Esx 3.5 "esxtop" you can see even more numbers - type e.g. "d", "F", "j" and reformat your putty screen.

You can see similar numbers for each vm - type "v", "F", "j"

CMDS/sec = ios/sec

DAVG/cmd = io latency in ms (1/1000 sec)

rd = read

wr = write

Re: Open inofficial storage performance thread

christianZ — Thu, 20 Nov 2008 07:49:00 GMT

Mnemonic

asp

thanks for your test results.

asp - I would first check the results with "esxtop' as I mentioned in the post above. Then you can try e.g. to clone one vm - what throughput can you see then?

By seq. read you should reach 90-100 MB/sec, I think. The array has power - the ios numbers by random tests are very high. I would check the /var/log file for any warnings/errors first.

Have you redundant controller there?

Re: Open inofficial storage performance thread

christianZ — Thu, 20 Nov 2008 07:39:13 GMT

Have you try with MS Initiator ? What the size of the lun?

Yes - ca. the same numbers; the lun size is 200 GB w/ thin provisioning.

Re: Open unofficial storage performance thread

christianZ — Thu, 20 Nov 2008 07:35:47 GMT

Well you can always verify your results with "esxtop" running directly on your Esx COS.

Then type "d" for disks and "F" / "j" (works on 3.5) - the putty window could be to small - change the character sizes.

Re: Open unofficial storage performance thread

acnsys — Wed, 19 Nov 2008 17:26:11 GMT

I have started the test but get weird values. I have a vmdk attached to the VM not formatted that appears as a physical drive in IOmeter. Besides as others i cant find the average -Av. IOs/sek-------Av. MB/sek so i am giving the Total IOs/sec and Total MB/sec (is that the right value?)

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM Win2003 3GB RAM/1 VCPU
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: ProLiant BL460c G1, 16GB RAM; 2x XEON 5355, 2,66 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EMC CX3-20f / ? Disks / R10

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........____0.0445______..........____21989.80______........._______687.18___

RealLife-60%Rand-65%Read......_____4.0389_____..........____247.50______.........______1.93____

Max Throughput-50%Read..........___5.7305_______.........._____10212.26_____.........___319.13_______

EXCEPTIONS: CPU Util.-XX%;

Re: Open unofficial storage performance thread

Mnemonic — Wed, 19 Nov 2008 10:05:40 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM on ESX 3.5.0 Update 3
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: HP Proliant DL580C G5, 48GB RAM; 4x XEON E7350 2,93 GHz Quad-Core
SAN Type: HP EVA 8100 / Disks: 4GB FC 300GB 15k / RAID LEVEL: Raid5 / 112 Disks / Qlogic QLA2432 4Gbit FC HBA

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read.......______5___..........___10854__........___339____

RealLife-60%Rand-65%Read......______5___..........____9018__........____70____

Max Throughput-50%Read........._____20___..........____2878__........____90____

Random-8k-70%Read................______5___..........____8830__........____69____

##################################################################################

Yes your throughput does seem odd.. Try setting "Net.TcpIpHeapSize = 30" in your advanced settings on the ESX server. Dont forget to reboot.

Re: Open unofficial storage performance thread

asp24 — Tue, 18 Nov 2008 14:20:48 GMT

SERVER TYPE: 2003 VM on 3.5.0
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Supermicro Dual Opteron Quadcore, 2GHz, 16GB RAM
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Infortrend S16E-G1130 ISCSI SAN, 16x 300GB 15k SAS, RAID 10
Only single gigabit link used. Sw Iscsi init. No jumbo frames

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........_____32_____.........._____1813_____........._____56,7_____ 21% CPU

RealLife-60%Rand-65%Read......_____12_____.........._____4401_____........._____34,4_____ 41% CPU

Max Throughput-50%Read.........._____30_____..........____1952______........._____61_____ 24% CPU

Random-8k-70%Read................._____12_____..........____4160______.........____32,5______ 42% CPU

##################################################################################

The performance is soooooo much better than my old Open-E SAN (in real life + random results in IOMETER),
but for some reason the IOMETER Max Throughput test results are very low.
Using ESXPRESS for backup I can read 100MB/s + from 2 servers at the same time (I guess more is possible)

Any ideas about the low Max Throughput test results?

Re: Open unofficial storage performance thread

dlima — Fri, 14 Nov 2008 19:52:29 GMT

I'm using Iometer 2006.07.27 with the config file provided but I can't see the Average I/O counter and the Average MB/sek, do I need to compute the average using (MB/Read + MB/Write)/2 ?, same thing for IO/s operations ?

Thanks in advance

David

Re: Open unofficial storage performance thread

Mnemonic — Mon, 27 Oct 2008 20:35:19 GMT

TABLE SAMPLE

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM on ESX 3.5.0 Update 3
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: HP Proliant DL580C G5, 16GB RAM; 2x XEON E7330 2,4 GHz Quad-Core
SAN Type: HP Eva 4400 / Disks: 4GB FC 136GB 15k / RAID LEVEL: Raid5 / 52+2 Disks / HP FC1142SR 4Gbit FC HBA

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........____5_____..........___10825__.........___338____

RealLife-60%Rand-65%Read......_____8____..........___6214___.........____49____

Max Throughput-50%Read..........____18____..........__2986____.........____93____

Random-8k-70%Read.................____8_____..........____5932__.........____46____

EXCEPTIONS: VMFS LUN

##################################################################################

Re: Open unofficial storage performance thread

Mnemonic — Mon, 27 Oct 2008 20:20:24 GMT

TABLE SAMPLE

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM on ESX 3.5.0 Update 3
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: HP Proliant DL580C G5, 16GB RAM; 2x XEON E7330 2,4 GHz Quad-Core
SAN Type: HP Eva 4400 / Disks: 4GB FC 136GB 15k / RAID LEVEL: Raid5 / 52+2 Disks / HP FC1142SR 4Gbit FC HBA

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........____5_____..........___10790__.........___337____

RealLife-60%Rand-65%Read......_____9____..........___5703___.........____45____

Max Throughput-50%Read..........____15____..........__3541____.........____111___

Random-8k-70%Read.................____9_____..........____5639__.........____44____

EXCEPTIONS: Continues Access Enabled VMFS LUN

##################################################################################

Re: Open unofficial storage performance thread

Mnemonic — Mon, 27 Oct 2008 20:06:27 GMT

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........____17_____..........___3359___.........___105____

RealLife-60%Rand-65%Read......_____13___..........___3841___.........____30____

Max Throughput-50%Read..........____19____..........__3073____........._____96___

Random-8k-70%Read.................____17_____..........____3103__.........____24____

EXCEPTIONS:

##################################################################################

Re: Open inofficial storage performance thread

darren.wardle50 — Mon, 29 Sep 2008 10:16:29 GMT

Thanks, I tried that but I couldn't find the values in excel.

Re: Open inofficial storage performance thread

larstr — Mon, 29 Sep 2008 07:20:57 GMT

Darren,

It's easiest to read the results if you open the .cvs file inside a spreadsheet application such as Excel, OpenOffice Calc, Lotus 123 or Visicalc.

Your results would be like this:

RealLife-60%Rand-65%Read......71.98...................633..............____4.95__

Max Throughput-50%Read........13.38...................4473.............___139.78__

Random-8k-70%Read.............66.73.......................778............_____6.08__

EXCEPTIONS: CPU Util. 57%-44%-25%-28%;

-##################################################################################

Re: Open inofficial storage performance thread

christianZ — Mon, 15 Sep 2008 19:25:41 GMT

chrisfmss wrote:
I use the IOmeter config file provided in this tread. What do you mean for the test file ?

During the first test a 4 GB test file will be created. Please check on c: or d: if you have one.

Sorry for my mistrust .

Re: Open inofficial storage performance thread

chrisfmss — Mon, 15 Sep 2008 18:47:37 GMT

I use the IOmeter config file provided in this tread. What do you mean for the test file ?

Re: Open inofficial storage performance thread

christianZ — Mon, 15 Sep 2008 18:43:05 GMT

Woh. I'm still a little sceptical - that is better than the DS4800. Are you sure your testfile is 4GB and have you verified your results with e.g. "esxtop / d" direct on the console.

Re: Open inofficial storage performance thread

chrisfmss — Mon, 15 Sep 2008 18:41:14 GMT

My test was 10 GB. Can this make a difference ?

Re: Open inofficial storage performance thread

christianZ — Mon, 15 Sep 2008 18:38:18 GMT

chrisfmss wrote:
Have you try with MS Initiator ?

There are 2 iscsi hbas (Qlogic) and the ms iscsi initiator is used only for multipathing/load balancing

What the size of the lun?

200 GB

Re: Open inofficial storage performance thread

chrisfmss — Sat, 13 Sep 2008 12:58:08 GMT

Have you try with MS Initiator ? What the size of the lun?

Re: Open inofficial storage performance thread

bjselman — Sat, 13 Sep 2008 10:18:10 GMT

I am out of the office all week, September 15-19, attending VMworld 2008. Please contact Scott Holmes by email: smholmes@travelhost.com, or extension 115.

Thanks,
BJ

Re: Open inofficial storage performance thread

christianZ — Sat, 13 Sep 2008 10:15:48 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS / PRODUCTION INFRASTRUCTURE
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: PH. Win 2003/64 sp2
CPU TYPE / NUMBER: PH. CPU / 2
HOST TYPE: Dell PE2950, 16GB RAM; 2x XEON X5450, 3,0 GHz, QC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS5000 x 1 / 14+2 SAS15k / R10 (FW 3.3.1 w/ thinprov.)
SAN TYPE / HBAs : iSCSI, QLA4050 HBA X 2, LB W/O JUMBO

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........____9,7____..........___6195___.........___194____

RealLife-60%Rand-65%Read......___12,8____..........__4032___.........___31,5____

Max Throughput-50%Read..........____9,7____...........___6152____.........___192____

Random-8k-70%Read.................____13,4___..........___3975___.........____31____

EXCEPTIONS: CPU Util. <10%,

##################################################################################

Could not reach the numbers of PS5000 with sata (new FW 4.0.1 here). Is that the new Fw that makes such difference (or maybe jumbos)? I wonder.

Re: Open inofficial storage performance thread

chrisfmss — Fri, 12 Sep 2008 19:35:24 GMT

I use MS initiator included in Hit Kit 3.1.1 with MPIO on Windows 2003 SP2 64 bits. Can the switch make a difference ? I use 2 Cisco 3750-E in one stack. Jumbo frame and flowcontrol on. Another thing i see from your setup, you are using Raid 50. I am using the new firmware 4.0.1.

Re: Open inofficial storage performance thread

christianZ — Fri, 12 Sep 2008 19:26:45 GMT

RealLife-60%Rand-65%Read......___11.8___..........__4314.28__.........___33.71___

I wonder about your results here - that all with satas. I'm a little sceptical, if it possible with satas.

Have here one new box with SAS 15k disks, want to test it as well.

Re: Open inofficial storage performance thread

christianZ — Fri, 12 Sep 2008 18:01:11 GMT

Ok. I wrote that your results are not so good but now I can see the results from happyhammer (the same box cx3-40) and can see much more as by your tests (and think these are much more realistic numbers for such storage).

Maybe there is some configuration option to make it faster - should be possible, I think.

Re: Open inofficial storage performance thread

chrisfmss — Fri, 12 Sep 2008 17:05:54 GMT

There is my test with only one array.

SERVER TYPE: PHYS.
CPU TYPE / NUMBER: XEON 3.2GHz / 1
HOST TYPE: Proliant ML350 G3, 4GB RAM; 1x XEON, 3.2 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS5000E x 1 / (14+2) 1TB Disks SATA / R10
MS Initiator, flow control enable and jumbo frame enable. 2 x 1GB Nics

Max Throughput-100%Read........___9.4___..........__5091.55__.........___159.11___

RealLife-60%Rand-65%Read......___11.8___..........__4314.28__.........___33.71___

Max Throughput-50%Read..........___9.0____..........__5925.95__.........__185.19__

Random-8k-70%Read.................___12.2____..........__4186.12__.........__32.70___

EXCEPTIONS: CPU Util.-XX%;

##################################################################################

Re: Open inofficial storage performance thread

larstr — Fri, 12 Sep 2008 17:02:25 GMT

chrisfmss wrote:

This my result for my new environment. I will do another test later with juste one array.

Are these result good for an Equallogic Product?

It looks like you're having some timing problems. Your mixed IO results are twice as high as your max throughput tests and that's not normal behavior. Either you have a too little test file (so you only test against RAM instead of disk) or you have some timing problems. How much RAM are there on your SAN controllers?

Lars

Re: Open inofficial storage performance thread

wpatton — Fri, 12 Sep 2008 14:50:29 GMT

Chrisfmss,

You can do an easy compare and search of other results using this online doc: http://spreadsheets.google.com/ccc?key=p2IFgyUF_v5Jn-7QobgY9Fw

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

Re: Open inofficial storage performance thread

chrisfmss — Fri, 12 Sep 2008 14:45:28 GMT

This my result for my new environment. I will do another test later with juste one array.

Are these result good for an Equallogic Product?

SERVER TYPE: PHYS.
CPU TYPE / NUMBER: XEON 3.2GHz / 1
HOST TYPE: Proliant ML350 G3, 4GB RAM; 1x XEON, 3.2 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS5000E x 2 / 2 x (14+2) 1TB Disks SATA / R10
MS Initiator, flow control enable and jumbo frame enable. 2 x 1GB Nics

Max Throughput-100%Read........___9.54___..........__5542.56__.........___170.39___

RealLife-60%Rand-65%Read......____4.8___.........._10945.23__.........___85.51___

Max Throughput-50%Read..........___8.6____..........__5660.31__.........__176.88__

Random-8k-70%Read.................___4.5____.........._11656.56__.........__91.07___

EXCEPTIONS: CPU Util.-XX%;

##################################################################################

Re: Open inofficial storage performance thread

zbenga — Thu, 11 Sep 2008 02:35:39 GMT

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1001778

i had the same errors on Clariion it's to do with the kernel in linux guests, will try latest kernel

Re: Open unofficial storage performance thread

bjselman — Fri, 08 Aug 2008 13:05:12 GMT

I am out of the office today, Thursday, August 7, attending a virtualization conference. If you have an IT emergency, please contact Scott Holmes.

Thanks,
BJ

Re: Open unofficial storage performance thread

happyhammer — Fri, 08 Aug 2008 12:56:35 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell PE2950, 32GB RAM; 2x XEON 5450QUAD, 3.00 GHz,
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Clarion CX340/ 9 Disks / R5
VMFS: 344GB LUN, 1MB Block Size
SAN TYPE / HBAs : 4GB FC, QLogic QLA2432 Dual HBAs, Dual McData 4700 Switches

Max Throughput-100%Read........4.99..................11619...................363.1______

RealLife-60%Rand-65%Read......17.42..............._2865....................22.39________

Max Throughput-50%Read........4.18.................13868...................433.38________

Random-8k-70%Read.............17.21.................2948...................23.03_______

EXCEPTIONS: CPU Util.46%-36%-56%-35%;

Re: Open inofficial storage performance thread

cmanucy — Wed, 06 Aug 2008 15:39:40 GMT

The NFS via Linux has had mixed results in the past 6 months we've had it in use. While the Promise box works great with the MS iSCSI Initiator, I have a feeling they're not too into Linux development. From time to time, the array will go read-only on the Linux host, forcing me to reboot it (the Linux host) to get r/w access back to the box.

Your post prompted me to check for firmware updates - which they have. I'm on 2.26 now, and they've got 2.28SR3.1 out now. I'm going to try that as well.

To answer your other question, however, even ESX 3.5 won't login properly to the Promise box... keeps trying to establish connections to the portal, which fails over and over again. While this is slightly different behavior than I was seeing before, it's still a no-go for me at this time.

You can, however, do RDM's from a Windows-based guest (I've got 2 NICs dedicated to a separate iSCSI network on each ESX host just for this) ... this process seems to work well for us, we haven't experienced any issues to date (until you need to do a firmware upgrade!!)

----
Carter Manucy

Re: Open inofficial storage performance thread

surbahns — Wed, 06 Aug 2008 14:49:39 GMT

I was doing some research on this promise box and came across your post. I'm interested in getting this VTM610I and loading it with 16x 1TB drives. It would primarily be used for backups to disk before the backups make it to tape. But I would like to have the possilbities of mounting it as storage on my ESX farm if need me for a short term solution. Without using the NFS through a linux box as a front end have you had any more further luck in getting it to work directly with ESX? Maybe with ESX 3.5?

Thanks!

Re: Open unofficial storage performance thread

Atamido — Wed, 30 Jul 2008 20:09:47 GMT

The link should really be:
http://spreadsheets.google.com/ccc?key=p2IFgyUF_v5Jn-7QobgY9Fw

I think the issue is that only Christian and I were editing it, and it's a lot of information to enter. A whole lot. Perhaps if someone else would volunteer to add information, then it would get updated? If you would like to update it, just send me your email address and I'll add you.

Re: Open unofficial storage performance thread

wpatton — Wed, 30 Jul 2008 18:52:33 GMT

As the last entry on the spreadsheet is August 4th 2007, I am going to say that ChristianZ is no longer maintaining it. Which is a bummer as I think this is one of the best threads in VMTN!

Re: Open unofficial storage performance thread

saqp — Thu, 24 Jul 2008 08:49:53 GMT

Removed double entry.

Re: Open unofficial storage performance thread

saqp — Thu, 24 Jul 2008 08:38:22 GMT

Hi Guys

I figured i'd join in.
I did a test using Xtravirt XVS iscsi mirroring thing and thought you might find some performance data interesting.
for those of you that haven't had the oppotunity of reading up on XVS it's basically a setup where you have two hosts each running a VM using local disks that in turn serves up an ISCSI target.
Basically a use of local storage as mirrored ISCSI SAN much like Lefthand Networks.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS / XTRAVIRT XVS Virtual SAN
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM ON ESX 3.5 Upd. 1
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Proliant BL20p G2, 4GB RAM; 1x XEON , 3,0 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: Local / 2x36GB 10k / R1 (Fault tolerant Mirrored Virtual ISCSI SAN using local disks)
SAN TYPE / HBAs : iSCSI Software initiator

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........10.35..........5453.96.........170.44

RealLife-60%Rand-65%Read......40.01..........1240.08.........9.69

Max Throughput-50%Read..........6.88..........7636.94.........238.65

Random-8k-70%Read.................36.80..........1319.23.........10.31

EXCEPTIONS: CPU Util.-(lots)

##################################################################################

Regards.
/Stig.

Re: Open unofficial storage performance thread

VirtualNoitall — Wed, 09 Jul 2008 23:49:54 GMT

At the bottom of the first post in this thread is a link to a google docs spreadseet. If it were up to date it would be exactly what you are after.

Is anyone maintaining the spreadsheet?

Re: Open unofficial storage performance thread

stormlight — Mon, 23 Jun 2008 15:22:19 GMT

is there a way to search just this thread? I want to search for certain models and compare results. However when i do a search above it searches the whole ENT stratagy and planning .

Re: Open inofficial storage performance thread

christianZ — Sun, 18 May 2008 10:03:11 GMT

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS / PRODUCTION INFRASTRUCTURE
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM ON ESX 3.5 Upd. 1
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell PE2950, 16GB RAM; 2x XEON X5450, 3,0 GHz, QC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 3 / 14+2 SAS10k / R50 (FW 3.3.1 w/o thinprov.)
SAN TYPE / HBAs : iSCSI, QLA4050 HBA

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........__17______..........___3500___.........___110____

RealLife-60%Rand-65%Read......___20_____..........___2765___.........____22____

Max Throughput-50%Read..........____14____..........___4190___.........___131____

Random-8k-70%Read.................____22____..........___2566___.........____20____

EXCEPTIONS: VCPU Util. <22%, all tests running on prod. systems; storage units (3) cap. utilization > 50 %;
runnings machines: vms ca. 45/(ca. 100 configured), ph. ca. 5, in production since 03.2007
##################################################################################

Test running on stripped volume (3 members). As you can see here the performance is not so good
as by beginning (the first tests done on fresh system).But this behavior was expected -
should be similar on other systems also.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE OF RESULTS / PRODUCTION INFRASTRUCTURE
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM ON ESX 3.5 Upd. 1
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell PE2950, 16GB RAM; 2x XEON X5450, 3,0 GHz, QC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 1 / 14+2 SAS10k / R50 (FW 3.3.1 w/o thinprov.)
SAN TYPE / HBAs : iSCSI, QLA4050 HBA

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........__17______..........___3540___.........___110____

RealLife-60%Rand-65%Read......___36_____..........___1590___.........____12____

Max Throughput-50%Read..........____11____..........___5620___.........___175____

Random-8k-70%Read.................____39____..........___1468___.........____12____

EXCEPTIONS: VCPU Util. <22%, all tests running on prod. systems; storage unit (1) cap. utilization > 50 %;
runnings machines: vms ca. 45/(ca. 100 configured), ph. ca. 5, in production since 03.2007

##################################################################################

These tests are running on not stripped volume (only one member) - it is a large difference
to the first test on fresh system. I wonder.

Re: Open unofficial storage performance thread

dalepa — Tue, 20 May 2008 04:48:15 GMT

I took the NFSSTAT results from our Netapp 3070's running ~400 VMs each all using NFS... I found it interesting that over 50% of the blocks were 4K in size and writes account for more then 65% of the IOPS.

Do we need a 4K IOMeter test here?

What we really need is a tool that shows the IO workload of a given ESX host. Then we would be able to come up with a IOmeter test for better benchmarks.

Re: Open unofficial storage performance thread

larstr — Mon, 19 May 2008 08:18:20 GMT

If you open that results file in your favorite spreadsheet application and choose to import it comma separated, you can pick out the needed numbers and you'll get:

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM or PHYS.
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell PE6850, 16GB RAM; 4x XEON 51xx, 2,66 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 1 / 14+2 Disks / R50

##################################################################################
TEST NAME-------------------Av. Resp. Time ms------Av. IOs/sek-------Av. MB/sek------
##################################################################################

Max Throughput-100%Read........___63.63_.........._____922.20_.........____28.81__

RealLife-60%Rand-65%Read......___149.81_..........____444.75_........._____3.47__

Max Throughput-50%Read..........___34.00_..........____1742.76_.........____54.46__

Random-8k-70%Read.................__159.96__........._____347.11_.........______2.71__

EXCEPTIONS: CPU Util. 4%, 6%, 6%, 9%;

##################################################################################

Message was edited by: larstr
Wish there could have been a monospace font option in this editor

Re: Open unofficial storage performance thread

anh — Mon, 19 May 2008 07:35:07 GMT

How do i get an output like this ?

TABLE SAMPLE

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
TABLE oF RESULTS
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM or PHYS.
CPU TYPE / NUMBER: VCPU / 1
HOST TYPE: Dell PE6850, 16GB RAM; 4x XEON 51xx, 2,66 GHz, DC
STORAGE TYPE / DISK NUMBER / RAID LEVEL: EQL PS3600 x 1 / 14+2 Disks / R50

Max Throughput-100%Read........__________..........__________.........__________

RealLife-60%Rand-65%Read......__________..........__________.........__________

Max Throughput-50%Read..........__________..........__________.........__________

Random-8k-70%Read.................__________..........__________.........__________

EXCEPTIONS: CPU Util.-XX%;

##################################################################################

I only get the the resultsfile as the one attached.

Regards
Anders Hansen

http://blog.goit.dk