Random Data Performance

Many of you correctly brought up concerns with our use of Iometer with SandForce drives. SandForce's controller not only looks at how data is being accessed, but also the actual data itself (more info here). The more compressible the data, the lower the write amplification a SandForce drive will realize. While Iometer's random access controls how data is distributed across the drive, the data itself is highly compressible. For example, I took a 522MB output from Iometer and compressed it using Windows 7's built in Zip compression tool. The resulting file was 524KB in size.

This isn't a huge problem because most controllers don't care what you write, just how you write it. Not to mention that the sort of writes you do on a typical desktop system are highly compressable to begin with. But it does pose a problem when testing SandForce's controllers. What we end up showing is best case performance, but what if you're writing a lot of highly random data? What's the worst performance SandForce's controller can offer?

An upcoming version of Iometer will allow you to use 100% random data to test just that. Our own Ryan Smith compiled an early release of this version of Iometer so we could see how much of an impact purely random data has on Iometer performance. When the Use Random Data box is checked, the actual data Iometer generates is completely random. To confirm I tried compressing a 161MB output from Iometer with Use Random Data enabled:

The resulting file was actually slightly bigger. In other words, it could not be compressed. Perfect.

Iometer Performance Comparison - 4K Aligned, 4KB Random Write Speed
  Normal Data Random Data % of Max Perf
Corsair Force 100GB (SF-1200 MLC) 164.6 MB/s 122.5 MB/s 74.4%
OCZ Agility 2 100GB (SF-1200 MLC) 44.2 MB/s 46.3 MB/s 105%

With data that can't be compressed, the SF-1500 (or SF-1200 with 3.0.1 firmware) will drop from 164.6MB/s to 122.5MB/s. That's still faster than any other SSD except for Crucial's RealSSD C300. The Agility 2 and other SF-1200 drives running 3.0.5 shows no performance impact as it's already bound by the performance of its firmware. Since the rest of the Agility 2's performance is identical to the Force drive I'll only include one set of results in the table below:

Iometer Performance Comparison
Corsair Force 100GB (SF-1200 MLC) Normal Data Random Data % of Max Perf
4KB Random Read 52.1 MB/s 42.8 MB/s 82.1%
2MB Sequential Read 265.2 MB/s 212.4 MB/s 80.1%
2MB Sequential Write 251.7 MB/s 144.4 MB/s 57.4%

Read performance is also impacted, just not that much. Performance drops to around 80% of peak, which does put the SandForce drives behind Intel and Crucial in sequential read speed. Random read speed drops to Indilinx levels. Read speed is impacted because if we write fully random data to the drive there's simply more to read back when we need it, making the process slower.

Sequential write speed actually takes the biggest hit of them all. At only 144.4MB/s if you're writing highly random data sequentially you'll find that the SF-1200/SF-1500 performs worse than just about any other SSD controller on the market. Only the X25-M is slower. While the impact to read performance and random write performance isn't terrible, sequential performance does take a significant hit on these SandForce drives.

Random Read/Write Speed Overall System Performance using PCMark Vantage
POST A COMMENT

60 Comments

View All Comments

  • johnlewis - Wednesday, April 21, 2010 - link

    Thanks for another great article. I'm patiently waiting for a (decent) 512 GB SSD in my budget so I can just throw everything besides media files on it; 256 GB might work, if I wasn't so damn lazy. Plus, I'd rather have a half full 512 GB drive than a 90+% full 256 GB drive. Reply
  • retnuh - Wednesday, April 21, 2010 - link

    90%+ full 256gb, I hear ya. I've been digging over as many SSD reviews as I can in the last couple days. I WANT to replace my 256gb 5400rpm in my notebook, but just can't quite squeeze everything into 200gb. Reply
  • gadgetguy10 - Wednesday, April 21, 2010 - link

    I am waiting until the price is at least down to $1 per gigabyte for a decent ssd. I figure I can get by with about 128gb of space. Reply
  • retnuh - Thursday, April 22, 2010 - link

    my problem is the ~120gb of development VMs, I can't get rid of them, but since I'm in vmware all day a SSD would be heaven sent for general performance. I'm keeping things pretty slim at ~190gb out of 256gb, but that 200gb mark is just too tight. I'd buy a 300gb agility 2 today if it existed. Reply
  • 529th - Wednesday, April 21, 2010 - link

    I thought the OWC controllers were discovered to have the SF 1200 controllers?

    Also, can we get a review on the 50g Vertex LE that are selling at New Egg, Thanks
    Reply
  • Anand Lal Shimpi - Wednesday, April 21, 2010 - link

    Those numbers are from the older OWC Mercury which used a limited run of SF-1500. The newer drives going forward are SF-1200 based. I'll be phasing them out of our graphs as a result.

    Take care,
    Anand
    Reply
  • dmayes - Thursday, April 22, 2010 - link

    When are we going to see benchmarks on the new intel driver that's faster than microsoft's driver and it supports trim with raid and we shouldn't go off of just prices MSRP but actual newegg prices and maybe even have a low to high # for example Intel X25-M Mainstream SSDSA2MH160G2R5 ($400 - $489). Also include the 80gb version specially since its around $215 - $225 Reply
  • Impulses - Thursday, April 22, 2010 - link

    I agree on the prices... Street prices of OCZ's Indillix drives are much lower than that Corsair Nova for instance, since OCZ ALWAYS has $20-30 rebates going on their drives (and they're just cheaper to begin with)... OCZ's Nova equivalent, the Solid 2, is like $300 flat after MIR.

    Intel's newer SATA drivers don't enable TRIM in RAID, just w/RAID... You can have a SSD w/TRIM support and two HDD in RAID on the same controller with said drivers (something you couldn't do before), but you still can't RAID two SSD and retain TRIM support. AFAIK they didn't dramatically alter performance either but if you've got a link to tests that say otherwise I'd love to see it... I haven't bothered to install them yet.
    Reply
  • eaw999 - Thursday, April 22, 2010 - link

    but you do have to admit it is strange that imsm 8.9 was used for the testbed instead of irst 9.6. 8.9 doesn't support trim at all! one has to wonder how this might affect (or not) the benchmark scores. Reply
  • dmayes - Thursday, April 22, 2010 - link

    "The SandForce SF-1200 controller used in the A-DATA S599 with Intel’s latest RST 9.6 drivers is the fastest 2.5 inch solid state drive for Windows users at this time. This combination is able to outperform every other drive we have tested to date in all around performance." Source tweaktown but they compared it to another 1200 drive instead of using the same ssd with both drivers. This is what intel says "Is there TRIM support for RAID configurations?

    Intel® Rapid Storage Technology 9.6 supports TRIM in AHCI mode and in RAID mode for drives that are not part of a RAID volume.

    A defect was filed to correct the information in the Help file that states that TRIM is supported on RAID volumes."
    Reply

Log in

Don't have an account? Sign up now