Performance vs. Transfer Size

All of our IOmeter sequential tests happen at a queue depth of 1, which is indicative of a light desktop workload. It isn't too far fetched to see much higher queue depths on the desktop. The performance of these SSDs also greatly varies based on the size of the transfer. For this next test we turn to ATTO and run a sequential write over a 2GB span of LBAs at a queue depth of 4 and varying the size of the transfers.

 

One of our main complaints about the firmware 1.4/1.5 for Vertex 4 was poor sequential read performance at small transfer sizes. Unfortunately, Agility 4 is even worse. With larger transfers, performance is okay, although not top notch. Once we go under 64KB IO size, the performance falls off considerably. For example at transfer size of 16KB, Agility 4 manages only 95.7MB/s, whereas Samsung 830 is reading at nearly 500MB/s. The good news is most client SSD performance isn't bound by small sequential transfers, but it's still a limitation of the drive.

 

Fortunately there is no such problem with sequential write performance. Vertex 4 is still faster but only at bigger transfer sizes and even then the difference is not too bad (~50MB/s).

AS-SSD Incompressible Sequential Performance AnandTech Storage Bench 2011
Comments Locked

41 Comments

View All Comments

  • CaedenV - Tuesday, September 4, 2012 - link

    The entire Agility/Vertex 2 line was pretty bad. The 3's had a rough start with several major firmware updates, but after that they have really upped their quality quite a lot. Not saying that they would hold a candle to the likes of Crucial, Intel, or Samsung, or that they should be trusted in a 'mission critical' role (because they shouldn't). But as budget drives for a home system, they are really not bad.

    I was burnt also with some of the early power supplies, which were absolutely trash. But after OCZ purchased PC Power and Cooling they really turned things around, and my most recent OCZ power supply has some of the most accurate voltages I have seen for anything near it's price range (and that is before rebate). Again, they are not for anything mission critical, but they are also not advertised for it, nor priced for it. The current ones at least work more than well enough for home use.

    At any rate, they are a budget company, and as such they are hit-and-miss. Complaining that a budget oriented company like OCZ does not compare on quality to the likes of a high-end company like Samsung is like comparing a Nissan to a BMW. They both sell cars, but they are not remotely in the same market.
  • celestialgrave - Tuesday, September 4, 2012 - link

    Did I miss it or was power not checked this time? Perhaps safe to assume very close to the Vertex 4 for power consumption?
  • hasseb64 - Wednesday, September 5, 2012 - link

    AnandTech, why are you given OCZ a FW variable in your tests?
    I thought that AnandTech was a serious hardware site?
    Do not let OCZ or any other sloppy company get a 2nd chance with a “FW at consumer development strategy”.
  • Superneato - Wednesday, September 5, 2012 - link

    I'm assuming the Agility 4 is the boot drive in this test. Exactly what method is used to TRIM the drive manually? I have the same question regarding how the drive was secure erased for this test.
  • Kristian Vättö - Wednesday, September 5, 2012 - link

    I just use Windows' Disk Management to format the drive, which will TRIM all user accessible LBAs.

    As for secure erase, I use Linux and its Terminal (basic hdparm commands).
  • Superneato - Wednesday, September 5, 2012 - link

    You can format the boot drive from inside the booted OS? I must be missing something or do you have the test SSD installed as a secondary drive?

    Thanks for the fast reply.
  • Kristian Vättö - Thursday, September 6, 2012 - link

    All SSDs are tested as secondary drives. Using it as a boot drive can create more variables (the OS may be reading/writing in the background, which would affect the results).
  • poccsx - Thursday, September 6, 2012 - link

    I still can't figure out who makes the fastest SSD, they all seem to have their strengths and weaknesses.
  • albtocxhtrqm - Thursday, September 6, 2012 - link

    nice review but when I look for a ssd i look for price and reliability and then speed. I don't see the reliability factor tested here.
    Why not a reliability index based on reviews on 20. 50. 100 main websites like amazon, newegg, tigerdirect etc..

    For instance:
    OCZ Agility 512GB newegg.ca, september 06, 2012, 38% 1 star

    The more websites the more reliable this index would be.
  • Eryxx - Thursday, October 11, 2012 - link

    I did it, I succumbed to my inner cheapness and saved a few $$ by buying an OCZ Petrol 128G ~7 months ago.

    Sure, it was faster than my WD Cav. Black 640GB, but from Day 1, the SSD gave me issues. I had to back it up weekly, do a secure erase using the OCZ Linux tools, then reimage the drive to keep it running correctly. If I let it go longer than a week without doing this, it would begin randomly locking up the system. Still, when it worked, it was much faster, so I suffered through.

    One day, the Petrol simply could not be detected by the BIOS, like it was just gone. Couldn't do anything with it, even the Linux tools could no longer detect it, so RMAd back to OCZ.

    Roughly 10 days after the Petrol failure, I received a brand-new Agility 4 128GB. Oh, joyous day, right? No.

    Right out of the package, my BIOS cannot see this drive. I've tried every procedure recommended by OCZ and dozens of users on their forums, this thing will not show up in AHCI mode, period. It WILL appear in IDE mode, and even in AHCI it DOES show up in Win Explorer and Device Manager once you boot with another drive.

    Yes, I've updated the BIOS, isolated the drive so it's connected all alone, new SATA cable, different ports, reflashed the SSD firmware, blah, blah, blah, this thing just won't detect in my system in AHCI. Will yet another drive fare any better? I highly doubt it, but I have little recourse but to try, or possibly wait for another firmware revision which might solve this.

    If you check the OCZ forums, there are a lot of other posts with similar issues on the Vx4/Ag4 line, so I don't seem to be alone in my misery.

    OCZ for your next SSD? Caveat emptor, my friend.

Log in

Don't have an account? Sign up now