Intel SSD 750 PCIe SSD Review: NVMe for the Client
by Kristian Vättö on April 2, 2015 12:00 PM ESTMixed Random Read/Write Performance
Mixed read/write tests are also a new addition to our test suite. In real world applications a significant portion of workloads are mixed, meaning that there are both read and write IOs. Our Storage Bench benchmarks already illustrate mixed workloads by being based on actual real world IO traces, but until now we haven't had a proper synthetic way to measure mixed performance.
The benchmark is divided into two tests. The first one tests mixed performance with 4KB random IOs at six different read/write distributions starting at 100% reads and adding 20% of writes in each phase. Because we are dealing with a mixed workload that contains reads, the drive is first filled with 128KB sequential data to ensure valid results. Similarly, because the IO pattern is random, I've limited the LBA span to 16GB to ensure that the results aren't affected by IO consistency. The queue depth of the 4KB random test is three.
Again, for the sake of readability, I provide both an average based bar graph as well as a line graph with the full data on it. The bar graph represents an average of all six read/write distribution data rates for quick comparison, whereas the line graph includes a separate data point for each tested distribution.
The SSD 750 does very well in mixed random workloads, especially when compared to the SM951 that is slower than most high-end SATA drives. The performance scales quite nicely as the portion of writes is increased.
 |
|||||||||
Mixed Sequential Read/Write Performance
The sequential mixed workload tests are also tested with a full drive, but I've not limited the LBA range as that's not needed with sequential data patterns. The queue depth for the tests is one.
In mixed sequential workloads, however, the SSD 750 and SM951 are practically indentical. Both deliver excellent performance at 100% reads and writes, but the performance does drop significantly once reads and writes are mixed. Even with the drop, the two push out 400MB/s whereas most SATA drives manage ~200MB/s, so PCIe certainly has a big advantage here.
 |
|||||||||
Facebook
Twitter
RSS
132 Comments
View All Comments
perula - Thursday, April 9, 2015 - link
[For Sell] Counterfeit Dollar(perula0@gmail.com)Euro,POUNDS,PASSPORTS,ID,Visa Stamp.Email/ perula0@gmail.com/
Text;+1(201) 588-4406
Greetings to everyone on the forum,
we supply perfectly reproduced fake money with holograms and all security features available.
Indistinguishable to the eye and to touch.
also provide real valid and fake passports for any country
delivery is discreet
We offer free shipping for samples which is 1000 worth fake as MOQ
oddbjorn - Tuesday, April 14, 2015 - link
I just recieved my 750 yesterday and soon found myself slightly bummed out by the lacking NVMe BIOS-support in my ASUS P8Z77-V motherboard. I managed to get the drive working (albeit non-bootable) by placing it in the black PCIe 2.0 slot of the mainboard, but this is hardly a long term solution. I posted a question to the https://pcdiy.asus.com/ website regarding possible future support for these motherboards and this morning they had publised a poll to check the interest for BIOS/UEFI-support for NVMe's. Please vote here if you (like me) would like to see this implemented! https://pcdiy.asus.com/2015/04/asus-nvme-support-p...Elchi - Wednesday, April 15, 2015 - link
If you are a happy owner of an older ASUS MB (z77, x79, z87) please vote for NVme support !http://pcdiy.asus.com/2015/04/asus-nvme-support-po...
iliketoprogrammeoo99 - Monday, April 20, 2015 - link
hey, this drive is now on preorder at amazon!http://amzn.to/1DDKwoI
only $449 on amazon.
vventurelli74 - Monday, May 4, 2015 - link
Lets say I had an Intel 5520 Chipset based computer that has multiple PCIe 2.0 Slots. I would be able to get almost the maximum read performance (Since PCIe 2.0 is 500MB/s per 1X, 4X = 2000MB/s, which is exciting on an older computer. I am curious as to if this would be a bootable solution on my desktop though. With 12 Cores and 24 Threads, this computer is far from under-powered, and it would be nice to breath life into this machine, but the BIOS would have no NVMe support that I can think of. I know it has Intel SSD support, but this is from a different era. I wish someone could confirm that this either will, or will not be bootable on non-MVMe mobo's. I am getting conflicting answers.vventurelli74 - Monday, May 4, 2015 - link
Nevermind, finally found the requirements that this drive will not be bootable on on NVMe machines, whats more is even using it as a 'secondary' drive requires UEFI apparently. My computer wouldn't be able to use this card at all? That would suck.xyvyx2 - Friday, May 8, 2015 - link
Great review!Kristian, any chance you have two of these drives in the same machine & you could test RAID0 performance? I'm running into some slow read performance when using two Samsung PCIe drives in a Dell server w/ a RAID1 or RAID0 config. It's not like regular bottlenecking where you hit a performance cap, but where transfer rate drops down to ~ 1/5th the speed at a lower xfer rate.
I thought this was just a Storage Spaces problem, but the same holds true w/ regular windows software raid. I got up to about 4,200 MB/sec, then it tanked. I then ran two simultaneous ATTO tests on two of the drives and they both behaved normally & peaked at 2,700 MB/sec... so I don't think I'm hitting a PCIe bus limitation... I think it's all software.
I posted more detail on Technet here:
https://social.technet.microsoft.com/Forums/en-US/...
shadowfang - Saturday, September 26, 2015 - link
How does the pcie card perform on a system without nvme?