Community
cancel
Showing results for 
Search instead for 
Did you mean: 
SGant
Beginner
2,536 Views

intel P3608

MB: SUPERMICRO X10DRH-iT

Proc: 2x INTEL Xeon E5-2620V3 2.4GHz 15MB 6C/12T

I have a question regarding Intel P3608.

i don't know if we get all the performance from the NVMe Intel P3608.

is there any other way to test the performance?

thx

CrystalDiskMark 5.2.1 x64 (C) 2007-2017 hiyohiyo

Crystal Dew World : http://crystalmark.info/

-----------------------------------------------------------------------

* MB/s = 1,000,000 bytes/s [SATA/600 = 600,000,000 bytes/s]

* KB = 1000 bytes, KiB = 1024 bytes

Sequential Read (Q= 32,T= 1) : 5395.325 MB/s

Sequential Write (Q= 32,T= 1) : 2015.354 MB/s

Random Read 4KiB (Q= 32,T= 1) : 462.633 MB/s [112947.5 IOPS]

Random Write 4KiB (Q= 32,T= 1) : 436.565 MB/s [106583.3 IOPS]

Sequential Read (T= 1) : 2259.450 MB/s

Sequential Write (T= 1) : 1899.885 MB/s

Random Read 4KiB (Q= 1,T= 1) : 37.511 MB/s [ 9158.0 IOPS]

Random Write 4KiB (Q= 1,T= 1) : 210.011 MB/s [ 51272.2 IOPS]

Test : 4096 MiB [X: 0.0% (0.2/1490.4 GiB)] (x5) [Interval=5 sec]

Date : 2017/01/15 16:32:53

OS : Windows Server 2016 Server Standard (full installation) [10.0 Build 14393] (x64)

Tags (1)
0 Kudos
37 Replies
idata
Community Manager
176 Views

The performance of the P3608 series of SSDs varies with their capacity. You did not specify which size P3608 you have, but looking at your CrystalDiskMark test results, it looks like it is the 1.6TB model.

CrystalDiskMark, and all simple storage drive benchmark programs are not able to test NVMe SSDs appropriately. Like Crystal, they only test with a maximum queue depth of 32, which is the maximum for the SATA/AHCI protocol. The P3608 uses the new NVMe protocol, which is faster and can work with many more outstanding IO requests than a SATA drive can. The testing configuration of Crystal limits the performance results of any NVMe SSD.

If you are comparing the IOPs results of the Crystal test, using a queue depth of 32, to the IOPs specifications of your P3608, which used a queue depth of 256 with a testing program called IOMeter, that is not a valid or accurate comparison.

The IOMeter program is the only free program available that has a chance of testing an NVMe SSD appropriately. Another free program called Anvil Storage Utilities, allows higher queue depth testing, but not to the limits of NVMe.

What makes you question the performance of your P3608? Do you have the Intel NVMe driver installed? There is a new version, just released: https://downloadcenter.intel.com/download/26451/Intel-SSD-Data-Center-Family-for-NVMe-Drivers https://downloadcenter.intel.com/download/26451/Intel-SSD-Data-Center-Family-for-NVMe-Drivers

SGant
Beginner
176 Views

I can not install "new" driver...https://downloadcenter.intel.com/download/26451/Intel-SSD-Data-Center-Family-for-NVMe-Drivers Download Intel® SSD Data Center Family for NVMe Drivers

Install LOG

2017:01:17 07:17:54:454:

2017:01:17 07:17:54:473: >>> Log start

2017:01:17 07:17:54:486: Single-instance mutex has been obtained

2017:01:17 07:17:54:499: Core version: 2.5.35

2017:01:17 07:17:54:511: Setup version: 2.0.20.0

2017:01:17 07:17:54:524: Command line: "C:\Users\Administrator.RCGASSIST\Downloads\1.8.0.1011\Installer\SetupNVME.exe"

2017:01:17 07:17:54:537: OS data: 10-0-3-0 64-bit

2017:01:17 07:17:54:550: System up time: 3382 sec

2017:01:17 07:17:54:563: Reboot pending: No

2017:01:17 07:17:54:576: Current UI language: 0409

2017:01:17 07:17:54:590: Language folder: C:\Users\ADMINI~1.RCG\AppData\Local\Temp\2\IIF9B4D.tmp

2017:01:17 07:17:54:659: Found C:\Users\ADMINI~1.RCG\AppData\Local\Temp\2\IIF9B4D.tmp\en-US\IntelCommon.dll. Lang 0409. Rank 3

2017:01:17 07:17:54:701: Loading language 0409

2017:01:17 07:17:54:716: Setup mode: Modify

2017:01:17 07:17:54:728: The Scsi device IOCTL execution succeeded. The RSTe NVMe driver is installed.

2017:01:17 07:17:54:740: Number of NVMe RAID volumes: 1

2017:01:17 07:17:54:752: The Scsi device IOCTL execution failed. Error = 55

2017:01:17 07:17:54:766: The RSTe NVMe driver was detected on the system.

2017:01:17 07:17:54:780: An RSTe NVMe RAID volume was detected on the system.

2017:01:17 07:17:54:793: E Win32 exception occurred!!

2017:01:17 07:17:54:805: Error code: 0x5209 (21001)

2017:01:17 07:17:54:818: Error text: ''

2017:01:17 07:17:54:830: Extended error text: 'The RSTe NVMe combined package is already installed. Installation of the standalone NVMe driver package may not continue.'

2017:01:17 07:17:54:844: Debug info: 'NVME.cpp (82)'

2017:01:17 07:17:54:857: Reading storyboard

2017:01:17 07:17:54:872: Page count: 7

2017:01:17 07:17:54:886: Added: Welcome

2017:01:17 07:17:54:898: Added: Modify

2017:01:17 07:17:54:915: Added: License

2017:01:17 07:17:54:928: Added: Confirm

2017:01:17 07:17:54:940: Added: Progress

2017:01:17 07:17:54:952: Added: Finish

2017:01:17 07:17:54:964: Added: Error

2017:01:17 07:17:54:976: Reading string map

2017:01:17 07:17:54:988: No string map present

2017:01:17 07:17:55:075: Showing page: Error

2017:01:17 07:18:30:349: Link: Launching 'C:\Users\Administrator.RCGASSIST\Intel\Logs'

idata
Community Manager
176 Views

Hello S.G.

We understand you have some questions in regards to the performance and benchmarking methods suggested for your http://www.intel.com/content/dam/www/public/us/en/documents/product-briefs/ssd-dc-p3608-brief.pdf Intel® SSD DC P3608 Series drive, which is currently being used in a RAID 0 array.

As @parsec correctly stated, the performance specifications for each SSD series do vary depending on drive capacity. Lower capacity equals higher random write IOPS, but lower sequential write speeds, and so forth.

The reason why our latest NVMe* drivers are failing to install is because your SSD is using the Intel® RSTe NVMe* RAID Driver instead. You may only use one or the other. And since your drive is set on a RAID array, the RSTe driver would be the correct one. You're also already using the latest of these.

As far as benchmarking for NVMe* drives, our suggestion is to use IOMeter® for Windows* and FIO® for Linux* OS.

- FIO®: http://freecode.com/projects/fio http://freecode.com/projects/fio.

 

- IOMeter®: https://sourceforge.net/projects/iometer/ https://sourceforge.net/projects/iometer/.

However, these applications are not the most user friendly, you may refer to the following pages for instructions:

- https://itpeernetwork.intel.com/how-to-benchmark-ssds-with-fio-visualizer/ How to benchmark SSDs with FIO Visualizer.

 

- http://www.intel.com/content/dam/www/public/us/en/documents/white-papers/performance-pcie-nvme-enter... Performance Benchmarking for PCIe* and NVMe* Enterprise Solid-State Drives.

For further IOMeter® instructions, you may also refer to the following document. This user guide was created with the Intel® SSD 750 Series in mind, but it applies equally for all Intel SSD Data Center NVMe* Series.

- http://www.intel.com/content/dam/support/us/en/documents/ssdc/hpssd/ssd-750/sb/intel_ssd_750_series_... Intel® SSD 750 Series Performance Evaluation Guide.

For the best performance, it's important to ensure your Intel® SSD DC P3608 Series is being connected to a PCIe* NVMe* Gen 3.0 x8 connector.

Note: Any links provided for third party tools or sites are offered for your convenience and should not be viewed as an endorsement by Intel® of the content, products, or services offered there. We do not offer support for any third party tool mentioned here.

Please let us know if this helps.

Best regards,

 

Carlos A.
SGant
Beginner
176 Views

this is my IOMETER result

SERVER TYPE: Windows Server 2016 VM running on Hyper-v

CPU TYPE / NUMBER: Xenon CPU E5-2600 v3 @ 2.40GHz (4 cores)

HOST TYPE: Supermicro, MB x10drh-it

STORAGE TYPE / DISK NUMBER / RAID LEVEL: NVMe, Intel P3608, RAID0

'Test Type,Test Description

0,$KB Random Reads

'Version

1.1.0

'Time Stamp

2017-01-18 10:58:57:163

'Access specifications

'Access specification name,default assignment

4KB Random Reads,0

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

4096,100,100,100,0,1,4096,0,

'End access specifications

'Results

'Target Type,Target Name,Access Specification Name,# Managers,# Workers,# Disks,IOps,Read IOps,Write IOps,MiBps (Binary),Read MiBps (Binary),Write MiBps (Binary),MBps (Decimal),Read MBps (Decimal),Write MBps (Decimal),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,0 to 50 uS,50 to 100 uS,100 to 200 uS,200 to 500 uS,0.5 to 1 mS,1 to 2 mS,2 to 5 mS,5 to 10 mS,10 to 15 mS,15 to 20 mS,20 to 30 mS,30 to 50 mS,50 to 100 mS,100 to 200 mS,200 to 500 mS,0.5 to 1 S,1 to 2 s,2 to 4.7 s,4.7 to 5 s,5 to 10 s, >= 10 s

ALL,All,4KB Random Reads,1,4,4,138737.333902,138737.333902,0.000000,541.942711,541.942711,0.000000,568.268120,568.268120,0.000000,138737.333902,0.000000,0.921459,0.921459,0.000000,0.921459,0.000000,205.087900,205.087900,0.000000,205.087900,0.000000,0,0,0,681935196160,0,166488085,0,0,-1,,,,,,,,,,0,16777216,32,64.801633,8.316838,56.486073,12.263503,0.959973,,30911.895893,2140.954270,8.532609,0.000000,0.025000,0,0,31600,19893633,98766118,41957138,5610766,209430,5983,1599,938,10368,0,481,31,0,0,0,0,0,0

MANAGER,WIN-CU432K6GC0C,4KB Random Reads,,4,4,138737.333902,138737.333902,0.000000,541.942711,541.942711,0.000000,568.268120,568.268120,0.000000,138737.333902,0.000000,0.921459,0.921459,0.000000,0.921459,0.000000,205.087900,205.087900,0.000000,205.087900,0.000000,0,0,0,681935196160,0,166488085,0,0,-1,1534118875738,0,1534118875738,0,2050879,0,2050879,0,12000251498,0,16777216,32,64.801633,8.316838,56.486073,12.263503,0.959973,10000000.000000,30911.895893,2140.954270,8.532609,0.000000,0.025000,0,0,31600,19893633,98766118,41957138,5610766,209430,5983,1599,938,10368,0,481,31,0,0,0,0,0,0

PROCESSOR,CPU 0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,91.150929,6.386800,84.765406,49.000626,3.121134,10000000.000000,20059.550531,,,,

PROCESSOR,CPU 1,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,59.888810,9.760541,50.129547,0.010417,0.286462,10000000.000000,4369.630011,,,,

PROCESSOR,CPU 2,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,56.482517,8.912874,47.570920,0.027344,0.238284,10000000.000000,3657.092572,,,,

PROCESSOR,CPU 3,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,51.684278,8.207136,43.478420,0.015625,0.194013,10000000.000000,2825.622779,,,,

WORKER,Worker 1,4KB Random Reads,,,1,34637.666324,34637.666324,0.000000,135.303384,135.303384,0.000000,141.875881,141.875881,0.000000,34637.666324,0.000000,0.922692,0.922692,0.000000,0.922692,0.000000,205.087900,205.087900,0.000000,205.087900,0.000000,0,0,0,170254462976,0,41566031,0,0,-1,383526343671,0,383526343671,0,2050879,0,2050879,0,12000240031,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,7842,4961200,24651359,10458355,1428799,53589,1502,438,227,2592,0,97,31,0,0,0,0,0,0

DISK,E: "New Volume",,,,,34637.666324,34637.666324,0.000000,135.303384,135.303384,0.000000,141.875881,141.875881,0.000000,34637.666324,0.000000,0.922692,0.922692,0.000000,0.922692,0.000000,205.087900,205.087900,0.000000,205.087900,0.000000,0,0,0,170254462976,0,41566031,0,0,-1,383526343671,0,383526343671,0,2050879,0,2050879,0,12000240031,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,7842,4961200,24651359,10458355,1428799,53589,1502,438,227,2592,0,97,31,0,0,0,0,0,0

WORKER,Worker 2,4KB Random Reads,,,1,34720.521939,34720.521939,0.000000,135.627039,135.627039,0.000000,142.215258,142.215258,0.000000,34720.521939,0.000000,0.920501,0.920501,0.000000,0.920501,0.000000,191.786800,191.786800,0.000000,191.786800,0.000000,0,0,0,170661535744,0,41665414,0,0,-1,383530732719,0,383530732719,0,1917868,0,1917868,0,12000226861,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,8036,4988833,24734924,10481989,1394687,52146,1464,387,228,2592,0,128,0,0,0,0,0,0,0

DISK,E: "New Volume",,,,,34720.521939,34720.521939,0.000000,135.627039,135.627039,0.000000,142.215258,142.215258,0.000000,34720.521939,0.000000,0.920501,0.920501,0.000000,0.920501,0.000000,191.786800,191.786800,0.000000,191.786800,0.000000,0,0,0,170661535744,0,41665414,0,0,-1,383530732719,0,383530732719,0,1917868,0,1917868,0,12000226861,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,8036,4988833,24734924,10481989,1394687,52146,1464,387,228,2592,0,128,0,0,0,0,0,0,0

WORKER,Worker 3,4KB Random Reads,,,1,34664.826822,34664.826822,0.000000,135.409480,135.409480,0.000000,141.987131,141.987131,0.000000,34664.826822,0.000000,0.921984,0.921984,0.000000,0.921984,0.000000,192.005800,192.005800,0.000000,192.005800,0.000000,0,0,0,170388127744,0,41598664,0,0,-1,383532822299,0,383532822299,0,1920058,0,1920058,0,12000251498,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,7682,4956629,24667162,10506454,1404004,51720,1673,374,246,2592,0,128,0,0,0,0,0,0,0

DISK,E: "New Volume",,,,,34664.826822,34664.826822,0.000000,135.409480,135.409480,0.000000,141.987131,141.987131,0.000000,34664.826822,0.000000,0.921984,0.921984,0.000000,0.921984,0.000000,192.005800,192.005800,0.000000,192.005800,0.000000,0,0,0,170388127744,0,41598664,0,0,-1,383532822299,0,383532822299,0,1920058,0,1920058,0,12000251498,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,7682,4956629,24667162,10506454,1404004,51720,1673,374,246,2592,0,128,0,0,0,0,0,0,0

WORKER,Worker 4,4KB Random Reads,,,1,34714.318817,34714.318817,0.000000,135.602808,135.602808,0.000000,142.189850,142.189850,0.000000,34714.318817,0.000000,0.920662,0.920662,0.000000,0.920662,0.000000,191.181100,191.181100,0.000000,191.181100,0.000000,0,0,0,170631069696,0,41657976,0,0,-1,383528977049,0,383528977049,0,1911811,0,1911811,0,12000228557,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,8040,4986971,24712673,10510340,1383276,51975,1344,400,237,2592,0,128,0,0,0,0,0,0,0

DISK,E: "New Volume",,,,,34714.318817,34714.318817,0.000000,135.602808,135.602808,0.000000,142.189850,142.189850,0.000000,34714.318817,0.000000,0.920662,0.920662,0.000000,0.920662,0.000000,191.181100,191.181100,0.000000,191.181100,0.000000,0,0,0,170631069696,0,41657976,0,0,-1,383528977049,0,383528977049,0,1911811,0,1911811,0,12000228557,0,16777216,32,,,,,,10000000.000000,,,,,,0,0,8040,4986971,24712673,10510340,1383276,51975,1344,400,237,2592,0,128,0,0,0,0,0,0,0

'Time Stamp

2017-01-18 11:18:57:322

'End Test

idata
Community Manager
176 Views

@Hello S.G.

 

 

The way that IOMeter* works, you will need to run two separate tests in order to get both read and write results. As it stands, it does seem like your SSD is under performing, but this could be a test configuration issue.

 

 

We can recommend running the tests in the following manner:

 

 

First Test ̶ Sequential Writes Test:

 

 

The first workload is the sequential writes test. Instead of using the whole drive, it only uses 1 worker at 32 outstanding I/Os per target over a span of 8GB. This is done by defining the Maximum Disk Size

 

 

For screenshots, refer the number on the section title to the pictures attached.

 

 

Setting the target Disk (1.jpg)
  1. Click on the Disk Target tab.
  2. Click on Worker 1.
  3. Click on the target drive
  4. Set Maximum Disk Size to 16777216 sectors.
  5. Leave Starting Disk Sector at 0.
  6. Set # of Outstanding I/Os to 32 per target.

Setting Access Specifications (2.jpg)

  1. Go to the Access Specifications Tab
  2. Under Global Access Specifications, click on New (A new window will open) (3a.jpg)
  3. Specify a name, ie. 128 KB Secuential Writes
  4. Set Transfer Request Size to 128 Kilobytes
  5. Set Percent/Sequential Distribution to 100% Sequential
  6. Set Percent Read/Write Distribution to 100% Write
  7. Press OK

Setting Up Test Duration (4.jpg)

  1. Click the Test Setup tab.
  2. Set for 20 minutes.

Running the Test (5.jpg)

  1. Click the Results Display tab.
  2. Set Last Update and Update Frequency to 1 second.
  3. Click the green flag to run the workload.
  4. Name and save the output file.
  5. Note the result under Total MBs per Second

--------------------------------------------------------------------------------------------

 

 

Second Test ̶ Sequential Reads Test:

 

 

The second workload is the sequential read workload. Just like the sequential writes workload, it only uses 1

 

worker at 32 outstanding I/Os per target, spans over 8GB, and is run for 20 minutes.

 

 

Setting the target Disk (1.jpg)
  1. Click on the Disk Target tab.
  2. Click on Worker 1.
  3. Click on the target drive
  4. Set Maximum Disk Size to 16777216 sectors.
  5. Leave Starting Disk Sector at 0.
  6. Set # of Outstanding I/Os to 32 per target.

Setting Access Specifications (2.jpg)

  1. Go to the Access Specifications Tab
  2. Under Global Access Specifications, click on New (A new window will open) (3b.jpg)
  3. Specify a name, ie. 128 KB Secuential Reads
  4. Set Transfer Request Size to 128 Kilobytes
  5. Set Percent/Sequential Distribution to 100% Sequential
  6. Set Percent Read/Write Distribution to 100% Read
  7. Press OK

Setting Up Test Duration (4.jpg)

  1. Click the Test Setup tab.
  2. Set for 20 minutes.

Running the Test (5.jpg)

  1. Click the Results Display tab.
  2. Set Last Update and Update Frequency to 1 second.
  3. Click the green flag to run the workload.
  4. Name and save the output file.
  5. Note the result under Total MBs per Second

Please let us know the results.

Best regards,

 

Carlos A.
SGant
Beginner
176 Views

here are my results

idata
Community Manager
176 Views

Hello S.G.,

I've reviewed the reports you've provided us, and I'm glad to inform you that your SSD is performing very well.

SSD performance is advertised based on results from tests ran on new/empty drives, under ideal conditions. Your overall hardware configuration, working temperature, amount of data currently stored on the drive, etc, will have an effect on your performance results. Results close to the 80% mark are considered normal.

Here are your results:

- Sequential Read: 5126 MB/s (Better than advertised for all capacities: 102% if 1.6TB or 4TB units, 113% if 3.2TB SSD).

 

- Sequential Write: 1952 MB/s (96% if 1.6TB SSD, 75% if 3.2TB unit).

As a point of comparison, you may review the following link:

- http://ark.intel.com/compare/88731,88734,88735 http://ark.intel.com/compare/88731,88734,88735

Please let us know if you have any further questions or concerns, as we'll be more than glad to continue assisting you.

Best regards,

 

Carlos A.
SGant
Beginner
176 Views

I am glad to hear that ...

what about random reads and writes.

Aren't random reads and writes more important for SQL server

What kind of test should i do just to be sure?

thanks

idata
Community Manager
176 Views

S.G.,

For this, we would need to run the tests again, but modifying only one step:

Under the "Setting Access Specifications" portion of the instructions, step 5 tells you to set the "Percent Random/Sequential Distribution". To test your random read/write performance, you'll simply need to make sure this option is set to 100% Random on each of the two tests. All other settings will remain the same.

When you get the results, instead of looking under the "Read/Write MBps (Decimal)" columns for Worker 1, you will need to look for the "Read/Write IOps" columns.

I've attached an example of what to look for in your results. I added color to some rows in order to make the file easier to read, but that's of course not necessary.

If you'd like for us to review and interpret these results for your SSD, feel free to post them on this thread.

Best regards,

 

Carlos A.
SGant
Beginner
176 Views

IoMeter results for Random R/W

idata
Community Manager
176 Views

Hello S.G.,

 

 

Testing for random read/writes is done a bit differently than we would for sequential results. This is due to the difference between random and sequential operations.

 

 

For random testing, you may follow the steps outlined in the http://www.intel.com/content/dam/support/us/en/documents/ssdc/hpssd/ssd-750/sb/intel_ssd_750_series_... Intel® SSD 750 Series Performance Evaluation Guide. Page 17 for Random Writes, Page 20 for the Random Reads tests.

 

 

You will need to make sure IOMeter* is configured with a Queue Depth (QD) of 32, Worker =4, and a 100% LBA range.

 

 

Best regards,

 

Carlos A.
SGant
Beginner
176 Views

IoMeter results for 4KB Random R/W

idata
Community Manager
176 Views

Hello S.G.,

 

 

The results from your latest set of tests are as follows:

 

 

Random writes: 127682 Write IOps (~85% of 150000)

 

Random reads: 131757 Read IOps (~15.5% of 850000)

 

 

Your random writes look great. In real world conditions, 85% is a great number. However, your random reads are too low, which to be frank, does not make sense considering all your other results have come back normal. Read speeds on solid state drives are expected to be much faster than write operations.

 

 

I will consult with our additional resources on possible causes and will let you know what we find.

 

 

Please continue monitoring this thread.

 

 

Best regards,

 

Carlos A.

 

SGant
Beginner
176 Views

Thank you for your info.

Maybe i'm doing something wrong regarding tests, but every tests i've done, the IOPS (random r/w) are pretty much the same

idata
Community Manager
176 Views

Hello S.G.,

 

 

Again, it's odd that your random reads are low when your sequential reads, as well as all other results, show a healthy drive performing as would be expected.

 

 

To rule any testing issues, let's run the test one last time. I'm aware I didn't mention it before, but this time, let's look at section 3 (mainly 3.2 and 3.3+) and make sure to follow all these recommendations.

 

 

Thank you for your patience.

 

 

Regards,

 

Carlos A.
SGant
Beginner
176 Views

in previous replies i've mentioned that all the test were run on virtual machine running on hyper-v server... i had problem running iometer on the server

now i've managed to run the test on "base" server, all the tests are 4KB random read,

i've disabled windows defender, windows updates, put the comp. in high performance, erased the disks

Intel P3608 1.6TB

here are the aprox results

1. single disk (800GB) without windows format = 400k IOPS

2. single disk (800GB) WITH windows format = 300k IOPS

3. RAID 0, created with Intel Rapid Storage Technology enterprise, without windows format; cannot do the test with Iometer (don't see the disk)

4. RAID 0, created with Intel Rapid Storage Technology enterprise, WITH windows format= 220k IOPS

5. RAID 0, created with windows disk management= 300k IOPS

SGant
Beginner
176 Views

installing Intel RTSe NVMe RAID Driver

https://downloadcenter.intel.com/download/25771/-Intel-Rapid-Storage-Technology-Enterprise-NVMe-Inte... Download Intel® Rapid Storage Technology Enterprise NVMe (Intel® RSTe NVMe) RAID Driver (4.5.0.2123)https://downloadcenter.intel.com/download/25771/-Intel-Rapid-Storage-Technology-Enterprise-NVMe-Inte... https://downloadcenter.intel.com/download/25771/-Intel-Rapid-Storage-Technology-Enterprise-NVMe-Inte...

or

https://downloadcenter.intel.com/download/26224/-Intel-Rapid-Storage-Technology-Enterprise-NVMe-Inte... Download Intel® Rapid Storage Technology Enterprise NVMe (Intel® RSTe NVMe) RAID Driver ( latest 4.5.0.2125 )

i always end up with driver version 4.5.0.2122...

idata
Community Manager
176 Views

Hello S.G.,

 

 

For benchmarking it's very important to follow the steps on the evaluation guide, as this is how we test the drives before posting the results to our ark.intel.com website. It's also very unlikely that a drive will pass only 3 of 4 benchmarking tests, as they are all closely related. I'm inclined to believe your SSD is performing as it should. We must be missing something when performing the random reads test, as the reading data is much simpler and faster on solid state drives than writing is (there are fewer steps involved).

 

 

As far as your Intel® RSTe driver version, we don't handle this directly (RST and RSTe are handled by our chipsets team). However, it's not uncommon for driver packages to contain different versions. Usually, the latest applicable version will be the one to install. Does the version also not match when you check from the user interface?

 

 

Best regards,

 

Carlos A.
SGant
Beginner
176 Views

1. why is there so big of diffference in Iometer tests

1. single disk (800GB) without windows format = 400k IOPS

2. single disk (800GB) WITH windows format = 300k IOPS

3. RAID 0, created with Intel Rapid Storage Technology enterprise, without windows format; cannot do the test with Iometer (don't see the disk)

4. RAID 0, created with Intel Rapid Storage Technology enterprise, WITH windows format= 220k IOPS

5. RAID 0, created with windows disk management= 300k IOPS

2. where can i check the version of my installed driver

regards, sg

idata
Community Manager
103 Views

Hello S.G.,

 

 

This is likely due to the difference in block sizes, which directly affects performance and benchmarking.

 

 

- http://www.tomshardware.com/reviews/ssd-dc-s3500-raid-performance,3613-8.html Six SSD DC S3500 Drives And Intel's RST: Performance In RAID, Tested - Mixing Block Sizes And Read/Write Ratios.

 

- http://www.linux-mag.com/id/7691/ Size Can http://www.linux-mag.com/id/7691/ Matter:http://www.linux-mag.com/id/7691/ Throughput Performance with a Disk-Based Journal.

 

 

As for the version, I was referencing to the possible difference between the driver and the RSTe versions. You're checking the driver version correctly, but I'd like to make sure that you have the latest RSTe software version installed. This can be checked in the Intel® RSTe Graphic User Interface (GUI) unless they've made changes to the recent release, it should be under the question sign icon on the taskbar.

 

 

- http://www.intel.com/content/dam/support/us/en/documents/ssdc/ssd-software/Intel_RSTe_NVMe_for_Windo... Intel® RSTe NVMe* for Windows Guide.

 

 

NOTE: Any links provided for third party tools or sites are offered for your convenience and should not be viewed as an endorsement by Intel® of the content, products, or services offered there. We do not offer support for any third party tool mentioned here.

 

 

Best regards,

 

Carlos A.
Reply