christianZ
Champion
Champion

New !! Open unofficial storage performance thread

Hello everybody,

the old thread seems to be sooooo looooong - therefore I decided (after a discussion with our moderator oreeh - thanks Oliver -) to start a new thread here.

Oliver will make a few links between the old and the new one and then he will close the old thread.

Thanks for joining in.

Reg

Christian

574 Replies
Khue
Enthusiast
Enthusiast

@ Roger,

I question your organizations choice in going after SATA on the VMware environment. It sounds like you have a substantially large SAN setup and your organization decided to skimp in a weird location. The SVC engines run off FC connections and therefore there is a significant cost of entry. You maybe saved like 200-300 dollars a drive going with SATA over FC drives but if the desire is to substitute a production, tier 1 environment with VMs I think I would have made an argument to go with FC drives in the 4700. I am not sure what types of drives the 5100 supports.

Your concern about your environment is justified I believe. I will run IO meter on my production environment sometime this week and post the results to compare. I have a similar setup and I will go into it when I make my post. I may have a question or two for you relating to how you ran IO meter specifically. I can say this, if I look at my SAN performance, on the ports that correspond to the VMware servers (I have 5) consume well over 2 gbps at times.

0 Kudos
sscheller
Contributor
Contributor

New Benchmark,

windows 2003 server with paralell buslogic.

h2bench from heise c´t test:

Kapazität: CHS=(2612/255/63), Jumbo Frames: Ohne Jumbo Frames:

41961780 Sektoren = 20489 MByte

Interface-Transferrate mit

Blockgröße 128 Sektoren bei 0.0% der Kapazität:

Sequenzielle Leserate Medium (ungebremst): 49948 KByte/s 56026 KByte/s

Sequenzielle Leserate Read-Ahead (Verzö:1.41 ms): 46321 KByte/s 56194 KByte/s

Wiederholtes sequenzielles Lesen ("Coretest"): 48551 KByte/s 56433 KByte/s

Dauertransferrate (Blockgröße: 128 Sektoren):

Lesen: Mittel 79740.8, 100404.9,

Min 8746.7, 27703.7,

Max 1093512.6 1109022.3

Zugriffszeit Lesen: Mittel 1.88, 1.84,

Min 0.01, 0.04,

Max 39.61 29.89

Zugriffszeit Lesen (<504 MByte): Mittel 3.38, 3.28,

Min 0.48, 0.55,

Max 26.69 32.30

Anwendungsprofil `Swappen': 11912.4 KByte/s 16346.3 KByte/s

Anwendungsprofil `Installieren': 27542.6 KByte/s 31614.3 KByte/s

Anwendungsprofil `Word': 62003.0 KByte/s 63495.0 KByte/s

Anwendungsprofil `Photoshop': 162709.4 KByte/s 125914.7 KByte/s

Anwendungsprofil `Kopieren': 101471.9 KByte/s 124241.1 KByte/s

Anwendungsprofil `F-Prot': 9800.2 KByte/s 12015.1 KByte/s

Gesamtergebnis: Anwendungsindex = 37.7 41.8

Test with iometer an vmware community standarts:

FLOW CONTROL!!!

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

TEST NAME--


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

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

Max Throughput-100%Read 54.228.400 ###### 34.728.489

RealLife-60%Rand-65%Read 306.433.999 195.146.580 1.524.583

Max Throughput-50%Read 301.445.890 199.367.148 6.230.223

Random-8k-70%Read 297.261.687 200.932.318 1.569.784

EXCEPTIONS: CPU Util.-XX%;

WITHOUT JUMBO AND FLOW CONTROL!!!

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

TEST NAME--


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

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

Max Throughput-100%Read 47.184.825 ###### 40.037.443

RealLife-60%Rand-65%Read 305.347.946 195.504.886 1.527.382

Max Throughput-50%Read 301.798.132 199.167.331 6.223.979

Random-8k-70%Read 298.059.851 200.717.557 1.568.106

EXCEPTIONS: CPU Util.-XX%;

WITHOUT JUMBO AND FLOW CONTROL!!!

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

TEST NAME--


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

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

Max Throughput-100%Read 47.184.825 ###### 40.037.443

RealLife-60%Rand-65%Read 305.347.946 195.504.886 1.527.382

Max Throughput-50%Read 301.798.132 199.167.331 6.223.979

Random-8k-70%Read 298.059.851 200.717.557 1.568.106

EXCEPTIONS: CPU Util.-XX%;

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

0 Kudos
Brechreiz
Contributor
Contributor

I'm rather new in VM Business, so please could someone rate the Results from my new Server Environement?

Jumbo Frames are enabled but I have no Information if Flow Control is enabled.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

TABLE oF RESULTS 1X VM WIN2008 64bit SP2 / ESX 4.0 ON Dell MD3000i via PC 5424

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM .

CPU TYPE / NUMBER: VCPU / 2 )JUMBO FRAMES, MPIO RR

HOST TYPE: Dell R610, 16GB RAM; 2x XEON 5540, 2,5 GHz, QC

ISCSI: VMWare iSCSI software initiator , Onboard Broadcom 5709 with TOE+ISCSI

STORAGE TYPE / DISK NUMBER / RAID LEVEL:Dell MD3000i x 1 / 6 Disks (15K 146GB / R10)

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

TEST NAME--


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

Max Throughput-100%Read........____14,55___..........____4133__.........____128,48____

RealLife-60%Rand-65%Read......_____22,69_.........._____2085__.........____16,92____

Max Throughput-50%Read.........._____14,13___.........._____4289__.........____134,04____

Random-8k-70%Read................._____21,7__.........._____2272__.........____17,75___

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

TABLE oF RESULTS 1X VM WIN2008 64bit SP2 / ESX 4.0 ON Dell MD3000i via PC 5424

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM .

CPU TYPE / NUMBER: VCPU / 2 )JUMBO FRAMES, MPIO RR

HOST TYPE: Dell R610, 16GB RAM; 2x XEON E5540, 2,5 GHz, QC

ISCSI: VMWare iSCSI software initiator , Onboard Broadcom 5709 with TOE+ISCSI

STORAGE TYPE / DISK NUMBER / RAID LEVEL:Dell MD3000i x 1 / 5+1 Disks (15K 146GB / R5)

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

TEST NAME--


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

Max Throughput-100%Read........____14,36___..........____4122__.........____128,68____

RealLife-60%Rand-65%Read......_____32,8_.........._____1067__.........____8,34____

Max Throughput-50%Read.........._____13,95___.........._____4339__.........____134,40____

Random-8k-70%Read................._____37,9__.........._____1066__.........____8,33___

0 Kudos
JohnADCO
Expert
Expert

Seems good for that few drives involved. I can't even find my results for my 3.5i environment, I was going to go compare them. I probably need to re-run the tests as well. Been some changes here and there, more VM's on the host.

0 Kudos
sscheller
Contributor
Contributor

Hi,

please tell me how you get such good performance?

Tower is same like IBm Ds3300.

Maybe my values are wrong.

How can i convert this big values ?

I am german people, with german excel.

Maybe there is a dot or comma problem.

What is your config?Tower?Switche?Iscsi settings?Raid etc?

0 Kudos
JohnADCO
Expert
Expert

Are you using ESX 4.0 or ESX 3.5?

They have improved the Initiator speed quite a bit in 4.0.....

0 Kudos
larstr
Champion
Champion

0 Kudos
sscheller
Contributor
Contributor

Hi,

we are using esx4 with software initiator iscsi.

Don´t know why performance is so poor.

0 Kudos
Brechreiz
Contributor
Contributor

@sscheller

Maybe the Raid-6 config is the Problem. You telling that you are from Germany. Me too. Am I right, that you posted your results also @ vmware-forum.de?

As far as I know is your DS 3300 similar to the MD3000i, so maybe this helps : http://www.delltechcenter.com/page/VMwareESX4.0andPowerVault+MD3000i

0 Kudos
sscheller
Contributor
Contributor

HI, thanks for your replay.

Yes iam from germany.

And right, we are using raid 6.

Yes did to.

My config looks like similar to your linked config.Our config only used one vmkernel with 1 IP for iscsi network for each esx server.The tower has got 4 nics, with ips from same subnet.Both esx server are redudant stacked on 2 switches.

But you only can ping from esx first nic from controller 1 and 2, because second nic from esx is offline.

This config looks realy interessting.

You can ping all 4 nics from tower with esx server.mh.

But which config is better?

Thanks

0 Kudos
Mnemonic
Enthusiast
Enthusiast

I dont understand how you get 134 MB pr. second through a 1 Gbit connection? The Teoretical Maximum is 1024/8 = 128 MB pr. second.

0 Kudos
sscheller
Contributor
Contributor

Hi,

can´t explain too.

These are the values from iometer.

Max Throughput-100%Read,1

'size,% of size,% reads,% random,delay,burst,align,reply

32768,100,100,0,0,1,0,0

'End access specifications

'Results

'Target Type,Target Name,Access Specification Name,# Managers,# Workers,# Disks,IOps,Read IOps,Write IOps,MBps,Read MBps,Write MBps,Transactions per Second,Connections per Second,Average Response Time,Average Read Response Time,Average Write Response Time,Average Transaction Time,Average Connection Time,Maximum Response Time,Maximum Read Response Time,Maximum Write Response Time,Maximum Transaction Time,Maximum Connection Time,Errors,Read Errors,Write Errors,Bytes Read,Bytes Written,Read I/Os,Write I/Os,Connections,Transactions per Connection,Total Raw Read Response Time,Total Raw Write Response Time,Total Raw Transaction Time,Total Raw Connection Time,Maximum Raw Read Response Time,Maximum Raw Write Response Time,Maximum Raw Transaction Time,Maximum Raw Connection Time,Total Raw Run Time,Starting Sector,Maximum Size,Queue Depth,% CPU Utilization,% User Time,% Privileged Time,% DPC Time,% Interrupt Time,Processor Speed,Interrupts per Second,CPU Effectiveness,Packets/Second,Packet Errors,Segments Retransmitted/Second

ALL,All,Max Throughput-100%Read,1,1,1,9822.969095,9822.969095,0.000000,306.967784,306.967784,0.000000,9822.969095,19.645218,6.104745,6.104745,0.000000,6.104745,50.900997,140.766502,140.766502,0.000000,140.766502,242.334670,0,0,0,96472530944,0,2944108,0,5888,500,,,,,,,,,,0,8000000,64,35.553227,5.958366,29.755145,3.925080,5.581055,,1767.833362,276.289100,711.567836,0.000000,0.000000

MANAGER,ADMIN-SC-PC,Max Throughput-100%Read,,1,1,9822.969095,9822.969095,0.000000,306.967784,306.967784,0.000000,9822.969095,19.645218,6.104745,6.104745,0.000000,6.104745,50.900997,140.766502,140.766502,0.000000,140.766502,242.334670,0,0,0,96472530944,0,2944108,0,5888,500,41877153951150,0,41877153951150,698312815751,327985949,0,327985949,564639782,698339939150,0,8000000,64,35.553227,5.958366,29.755145,3.925080,5.581055,2330000000.000000,1767.833362,276.289100,711.567836,0.000000,0.000000

PROCESSOR,CPU 0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,35.553227,5.958366,29.755145,3.925080,5.581055,2330000000.000000,1767.833362,,,,

WORKER,Worker 1,Max Throughput-100%Read,,,1,9822.969095,9822.969095,0.000000,306.967784,306.967784,0.000000,9822.969095,19.645218,6.104745,6.104745,0.000000,6.104745,50.900997,140.766502,140.766502,0.000000,140.766502,242.334670,0,0,0,96472530944,0,2944108,0,5888,500,41877153951150,0,41877153951150,698312815751,327985949,0,327985949,564639782,698339939150,0,8000000,64,,,,,,2330000000.000000,,,,,

DISK,C:,,,,,9822.969095,9822.969095,0.000000,306.967784,306.967784,0.000000,9822.969095,19.645218,6.104745,6.104745,0.000000,6.104745,50.900997,140.766502,140.766502,0.000000,140.766502,242.334670,0,0,0,96472530944,0,2944108,0,5888,500,41877153951150,0,41877153951150,698312815751,327985949,0,327985949,564639782,698339939150,0,8000000,64,,,,,,2330000000.000000,,,,,

'Time Stamp

2009-07-03 19:25:51:489

0 Kudos
lord_nibbler
Contributor
Contributor

Some interesting comparison between ESX 3.5 and ESX 4, and looks like the Server HBAs are hitting the maximum transfer rate (with the great Infortrend boxes and Max Throughput-50%Read)... amazing Smiley Happy

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

TABLE oF RESULTS

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM WinXP SP2 1GB RAM VM Version 7

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: ESX4 Build 164009, HP DL380 G6, 72GB RAM; 2x Intel Xeon X5570 (2.93 GHz), Quad Core, ISP2432-based 4Gb Fibre Channel Dual HBA

STORAGE TYPE / DISK NUMBER / RAID LEVEL: Infortrend S16F-R1430 Firmware 3.64P.07 / 8 x 300GB SAS 15k / RAID6

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

TEST NAME--


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

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

Max Throughput-100%Read......___4.95____......._11588.08__........._362.13___

RealLife-60%Rand-65%Read..___13.79____.......__3611.70__.........__28.22___

Max Throughput-50%Read........___3.14____.......__17588.21__........._549.63___

Random-8k-70%Read...............___16.24____.......__3082.23__.........__24.08___

EXCEPTIONS: CPU Util. 61% - 43% - 80% - 15%

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

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

TABLE oF RESULTS

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM WinXP SP2 1GB RAM VM Version 7

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: ESX4 Build 164009, HP DL380 G6, 72GB RAM; 2x Intel Xeon

X5570 (2.93 GHz), Quad Core, ISP2432-based 4Gb Fibre Channel Dual HBA

STORAGE TYPE / DISK NUMBER / RAID LEVEL: IBM DS4700 / 14 x 146GB FC 15k / RAID6

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

TEST NAME--


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

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

Max Throughput-100%Read......___5.32____......._10829.70__........._338.43___

RealLife-60%Rand-65%Read..___22.27____.......__2183.23__.........__17.06___

Max Throughput-50%Read........___13.14____.......__4447.26__........._138.98___

Random-8k-70%Read...............___19.39____.......__2273.20__.........__17.76___

EXCEPTIONS: CPU Util. 59% - 37% - 33% - xx%

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

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

TABLE oF RESULTS

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM WinXP SP2 1GB RAM VM Version 4

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: ESX3.5 Build 163429, HP DL380 G6, 72GB RAM; 2x Intel Xeon

X5570 (2.93 GHz), Quad Core, ISP2432-based 4Gb Fibre Channel Dual HBA

STORAGE TYPE / DISK NUMBER / RAID LEVEL: Infortrend S16F-R1430 Firmware 3.64P.07 / 8 x 300GB SAS 15k / RAID6

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

TEST NAME--


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

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

Max Throughput-100%Read......___5.03____......._11141.37__........._348.17___

RealLife-60%Rand-65%Read..___13.60____.......__3661.83__.........__28.61___

Max Throughput-50%Read........___3.66____.......__14973.10__........._467.91___

Random-8k-70%Read...............___16.05____.......__3128.80__.........__24.44___

EXCEPTIONS: CPU Util. 55% - 40% - 64% - 39%

0 Kudos
larstr
Champion
Champion

SERVER TYPE: VM with Windows Server 2008 x64 running on ESX 4.0 with paravirtualized scsi driver

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: DL380G6, 48GB RAM; 2x XEON 55xx,

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA 4400 / 24 Disks / R5

TEST NAME

Avg Resp. Time ms

Avg IOs/sec

Avg MB/sec

% cpu load

Max Throughput-100%Read

5.3

11249

351

30

RealLife-60%Rand-65%Read

11.6

3875

30

43

Max Throughput-50%Read

48.6

1130

35

22

Random-8k-70%Read

12.2

3800

30

40

0 Kudos
rb2006
Contributor
Contributor

Test

0 Kudos
ericmba
Contributor
Contributor

mate,

I agree with you, there is a lot I dont understand here. People running IO meter left and right and posting the results without regard for applications need for latency, network infrastr. etc. Oh well so long the results are not taken seriously Smiley Happy

Eric

0 Kudos
JaFF
Contributor
Contributor

Hey everyone,

Just been running through the tests on the same server and trying some different settings (with cache no bbwc, with 256MB cache + bbwc, with 512MB cache +bbwc)

Just hoping to see if these results are what you would expect.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

*TABLE OF RESULTS – SAS E Drive 100GB, 256MB, no BBWC – 0%write/100%read*

SERVER TYPE: VM – Windows 2003 STD

CPU TYPE / NUMBER: VCPU / 4

HOST TYPE: HP ML370 G6, 10GB RAM; 2x XEON 5540, 2,66 GHz

STORAGE TYPE / DISK NUMBER / RAID LEVEL: DAS / 7 Disks / R5

Test Name

Av. Resp Time ms

Av. IOs/sec

Av MB/sec

Max Throughput-100%Read

0.72

24650.70

770.33

RealLife-60%Rand-65%Read

49.31

1084.53

8.47

Max Throughput-50%Read

105.76

532.30

16.63

Random-8k-70%Read

43.57

1225.10

9.57

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

*TABLE OF RESULTS – SAS E Drive 100GB, 256MB, With BBWC – 50%write/50%read*

SERVER TYPE: VM – Windows 2003 STD

CPU TYPE / NUMBER: VCPU / 4

HOST TYPE: HP ML370 G6, 10GB RAM; 2x XEON 5540, 2,66 GHz

STORAGE TYPE / DISK NUMBER / RAID LEVEL: DAS / 7 Disks / R5

Test Name

Av. Resp Time ms

Av. IOs/sec

Av MB/sec

Max Throughput-100%Read

0.83

24681.12

771.28

RealLife-60%Rand-65%Read

21.52

1502.93

11.74

Max Throughput-50%Read

2.00

12915.28

403.60

Random-8k-70%Read

23.30

1401.65

10.95

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

TABLE OF RESULTS - SAS E Drive 100GB, 512MB, With BBWC - 50%write/50%read

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SERVER TYPE: VM - Windows 2003 STD

CPU TYPE / NUMBER: VCPU / 4

HOST TYPE: HP ML370 G6, 10GB RAM; 2x XEON 5540, 2,66 GHz

STORAGE TYPE / DISK NUMBER / RAID LEVEL: DAS / 7 Disks / R5

Test Name

Av. Resp Time ms

Av. IOs/sec

Av MB/sec

Max Throughput-100%Read

1.76

24748.11

773.37

RealLife-60%Rand-65%Read

21.53

1496.59

11.69

Max Throughput-50%Read

0.64

30740.11

960.62

Random-8k-70%Read

27.81

227.09

10.72

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Cheers.

J

0 Kudos
larstr
Champion
Champion

I did some more testing here. Most of it was done in Windows Server 2008 x64 with the default virtual LSI controller on different RAIDs and different number of disks. I also tested the paravirtual driver (on 2008) and Windows Server 2003r2 x32 for comparisons sake.

SERVER TYPE: Windows Server 2003r2 x32 VM with LSI Logic controller, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 23 Disks / RAID5

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.5

10831

338.47

38

RealLife-60%Rand-65%Read

10.8

4313

33.70

45

Max Throughput-50%Read

31.6

1822

56.95

17

Random-8k-70%Read

9.9

4613

36.04

47

SERVER TYPE: Windows Server 2008 x64 VM with LSI Logic SAS controller, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 23 Disks / RAID5

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.53

10865

339.5

34

RealLife-60%Rand-65%Read

10.27

4536

35.4

45

Max Throughput-50%Read

31.83

1817

56.78

17

Random-8k-70%Read

9.99

4443

34.7

49

SERVER TYPE: Windows Server 2008 x64 VM with LSI Logic SAS controller, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 23 Disks / RAID6

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.53

10871

339.73

33

RealLife-60%Rand-65%Read

11.5

3959

30.9

46

Max Throughput-50%Read

32.78

1792

56

18

Random-8k-70%Read

11.3

3808

29.76

50

SERVER TYPE: Windows Server 2008 x64 VM with paravirtual disk controller, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 23 Disks / RAID5

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.39

11140

348

30

RealLife-60%Rand-65%Read

10.15

4538

35.46

43

Max Throughput-50%Read

31.83

1816

56.75

14

Random-8k-70%Read

9.83

4436

34.65

46

SERVER TYPE: Windows Server 2008 x64 VM with LSI Logic, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 23 Disks / RAID10

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.5

10899

341.60

32

RealLife-60%Rand-65%Read

9.06

5649

44.14

39

Max Throughput-50%Read

10.68

4106

128.30

48

Random-8k-70%Read

8.84

5794

45.27

39

SERVER TYPE: Windows Server 2008 x64 VM with LSI Logic SAS controller, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 48 Disks / RAID5

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.52

10903

340.7

34

RealLife-60%Rand-65%Read

8.05

5884

46

46

Max Throughput-50%Read

33.96

1699

53.1

16

Random-8k-70%Read

8.24

5700

44.5

46

SERVER TYPE: Windows Server 2008 x64 VM with LSI Logic SAS controller, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 48 Disks / RAID6

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.51

10903

340.7

33

RealLife-60%Rand-65%Read

9.63

4878

38.1

44

Max Throughput-50%Read

40.01

1399

43.7

21

Random-8k-70%Read

9.35

4967

38.8

45

SERVER TYPE: Windows Server 2008 x64 VM with LSI Logic SAS controller, ESX 4.0

CPU TYPE / NUMBER: VCPU / 1

HOST TYPE: HP BL490c G6, 64GB RAM; 2x XEON E5540, 2,53 GHz, QC

STORAGE TYPE / DISK NUMBER / RAID LEVEL: HP EVA6400 / 48 Disks / RAID10

Test name

Avg resp time

Avg IO/s

Avg MB/s

Avg % cpu

Max Throughput-100%Read

5.51

10905

340.8

32

RealLife-60%Rand-65%Read

8.20

6366

49.7

39

Max Throughput-50%Read

9.31

5279

165

43

Random-8k-70%Read

7.81

6734

52.6

39

Lars

0 Kudos
HdeJongh
Contributor
Contributor

Geachte,

Tot maandag 24 augustus ben ik niet aanwezig. Deze mail wordt enkele keren per week uitgelezen.

Voor dringende vragen kunt u mijn collega`s bereiken via helpdesk@itcreation.nl<mailto:helpdesk@itcreation.nl> of via 010 282 1600.

Mvg,

Hans de Jongh

Weena 290

Postbus 819

3000 AV Rotterdam

The Netherlands

Phone: 010 - 282 16 00

Fax: 010 - 282 13 01

Mobile: 065 - 322 63 31

Email: hans.de.jongh@itcreation.nl

Internet: http://www.itcreation.nl<http://www.itcreation.nl/>;

Support: http://helpdesk.itcreation.nl<http://helpdesk.itcreation.nl/

0 Kudos
Mnemonic
Enthusiast
Enthusiast

.

0 Kudos