Performance Consistency

Performance consistency tells us a lot about the architecture of these SSDs and how they handle internal defragmentation. The reason we do not have consistent IO latency with SSDs is because inevitably all controllers have to do some amount of defragmentation or garbage collection in order to continue operating at high speeds. When and how an SSD decides to run its defrag or cleanup routines directly impacts the user experience as inconsistent performance results in application slowdowns.

To test IO consistency, we fill a secure erased SSD with sequential data to ensure that all user accessible LBAs have data associated with them. Next we kick off a 4KB random write workload across all LBAs at a queue depth of 32 using incompressible data. The test is run for just over half an hour and we record instantaneous IOPS every second.

We are also testing drives with added over-provisioning by limiting the LBA range. This gives us a look into the drive’s behavior with varying levels of empty space, which is frankly a more realistic approach for client workloads.

Each of the three graphs has its own purpose. The first one is of the whole duration of the test in log scale. The second and third one zoom into the beginning of steady-state operation (t=1400s) but on different scales: the second one uses log scale for easy comparison whereas the third one uses linear scale for better visualization of differences between drives. Click the dropdown selections below each graph to switch the source data.

For more detailed description of the test and why performance consistency matters, read our original Intel SSD DC S3700 article.

AMD Radeon R7 240GB
Default
25% Over-Provisioning

The IO consistency is very similar to the ARC 100 but the R7 is maybe slightly faster. Compared to the Vector 150 and Vertex 460 there is a small decrease in consistency as performance occassionally drops below 10K IOPS, but on average IOPS of 15-20K is excellent for a client drive. The same goes for IO consistency with 25% over-provisioning – the R7 is not as good as the Vector 150 and Vertex 460 but it is still one of the best performing client SSDs.

AMD Radeon R7 240GB
Default
25% Over-Provisioning

 

AMD Radeon R7 240GB
Default
25% Over-Provisioning

 

Introduction, The Drive & The Test AnandTech Storage Bench 2013
Comments Locked

54 Comments

View All Comments

  • Death666Angel - Friday, August 29, 2014 - link

    Agility 1 and Vertex 2 still going strong here!
  • mapesdhs - Saturday, August 30, 2014 - link

    The key is fw updates. The old bugs are fixed. Once done, they're fine.

    People act like OCZ was the only company to have issues, but even Intel and
    Samsung screwed up their SSD line at one point or other (FUD posters choose
    to forget Intel's 8MB bricked SSD issue).

    Ian.
  • bronan - Monday, November 9, 2015 - link

    Agreed on that Death666Angel i got 12 of them all running perfect, ofcourse one day they will die :D
    But this far super and still good performance for such OLD ssd ;)

    On topic i like the way AMD is going open in everything no faking or hiding facts like the competition, i hope AMD keeps up the good work they are doing. Teaming up with other good brands like Toshiba and such only makes both brands better. A shame they do not have a enterprise version else i would have considered them as well
  • zero2dash - Thursday, August 28, 2014 - link

    I'm sorry, did you just say Vertex/Agility were reliable? Is it opposite day or something?

    I owned a 30 GB Vertex. It was great for 10 months. Then it started throwing up chkdsk errors. Sanitary Erasing did nothing but "buy some time". I had it replaced several times under warranty. Then they put out a FW update that turned Vertex's into Vertex Pros (probably in an effort to look less terrible with mud on their faces). I flashed mine, sanitary erased it again, and sold it for peanuts (almost literally) and got a Crucial M4 SSD that has been rock solid for several years now.

    As far as I'm concerned, OCZ can close their doors for good. Or don't, I could care less because I'm never purchasing another OCZ product ever again.
  • Homeles - Thursday, August 28, 2014 - link

    Their Octane/Petrol failure rates were close to 50%.
  • ProfSparkles - Friday, August 29, 2014 - link

    At my company we go through many SSDs per year and by now we lost every single OCZ SSD (mostly Vertex 3) we bought so far and had to replace it, at first we use the warranty and replaced it with yet another OCZ but when those failed aswell we bought new Samsung SSDs (840 EVO) which was less expensive than driving to customers another time and replacing the failed drives. The Samsungs drives already tripled the OCZ drives lifetime.
    I personally own a first batch Vertex 2 which still works well but since the rev 2 of the Vertex 2 and Vertex 3 they went so far downhill its just ridiculous.
  • mapesdhs - Saturday, August 30, 2014 - link


    Blah blah. Fact is, none of that applies to the non-Sandforce models, and I've had no
    issues with the pile of V3 MXIs I bought; guess you were just unlucky.

    Ian.
  • dragonsqrrl - Thursday, August 28, 2014 - link

    ... I think he's referring to their SSD's.
  • Guspaz - Thursday, August 28, 2014 - link

    Bought a bunch of OCZ RAM. It failed (or at least some sticks did) and they got out of the RAM business so they invalidated the warranty. Convinced a friend to buy a Vertex 2 before they got their bad rep. It failed. He got a replacement. It failed. He got a replacement. It failed. He bought an Intel. It didn't fail.

    Sorry, but a company with such unreliable products getting bought out by Toshiba (who lied to me and refused to honour their warranty) just makes them even more of a "not with a ten foot pole" in my books.
  • Sparrowgryphon - Monday, March 7, 2016 - link

    I just had this SSD fail last night, it was about 4 months old. put it in a different PC will not show up and even stops the PC from booting.

Log in

Don't have an account? Sign up now