Random Read Performance

One of the major changes in our 2015 test suite is the synthetic Iometer tests we run. In the past we used to test just one or two queue depths, but real world workloads always contain a mix of different queue depths as shown by our Storage Bench traces. To get the full scope in performance, I'm now testing various queue depths starting from one and going all the way to up to 32. I'm not testing every single queue depth, but merely how the throughput scales with the queue depth. I'm using exponential scaling, meaning that the tested queue depths increase in powers of two (i.e. 1, 2, 4, 8...). 

Read tests are conducted on a full drive because that is the only way to ensure that the results are valid (testing with an empty drive can substantially inflate the results and in reality the data you are reading is always valid rather than full of zeros). Each queue depth is tested for three minutes and there is no idle time between the tests. 

I'm also reporting two metrics now. For the bar graph, I've taken the average of QD1, QD2 and QD4 data rates, which are the most relevant queue depths for client workloads. This allows for easy and quick comparison between drives. In addition to the bar graph, I'm including a line graph, which shows the performance scaling across all queue depths. To keep the line graphs readable, each drive has its own graph, which can be selected from the drop-down menu.

I'm also plotting power for SATA drives and will be doing the same for PCIe drives as soon as I have the system set up properly. Our datalogging multimeter logs power consumption every second, so I report the average for every queue depth to see how the power scales with the queue depth and performance.

Iometer - 4KB Random Read

Random read performance at small queue depths has never been an area where the Vector 180 has excelled in. Given that these are one of the most common IOs, it's an area where I would like to see improvement on OCZ's behalf.

Iometer - 4KB Random Read (Power)

Power consumption, on the other hand, is excellent, which is partially explained by the lower performance. 

Samsung SM951 512GB

Having a closer look at the performance data across all queue depths reveals the reason for Vector 180's poor random read performance. For some reason, the performance only starts to scale properly after queue depth of 4, but even then the scaling isn't as aggressive as on some other drives. 

Random Write Performance

Write performance is tested in the same way as read performance, except that the drive is in a secure erased state and the LBA span is limited to 16GB. We already test performance consistency separately, so a secure erased drive and limited LBA span ensures that the results here represent peak performance rather than sustained performance.

Iometer - 4KB Random Write

In random write performance the Vector 180 does considerably better, although it's still not the fastest drive around. 

Iometer - 4KB Random Write (Power)

Even though the random write performance doesn't scale at all with capacity, the power consumption does. Still, the Vector 180 is quite power efficient compared to other drives.

Samsung SM951 512GB

The Vector 180 scales smoothly across all queue depths, but it could scale a bit more aggressively because especially the QD4 score is a bit low. On a positive side, the Vector 180 does very well at QD1, though.

AnandTech Storage Bench - Light Sequential Performance
POST A COMMENT

89 Comments

View All Comments

  • zodiacsoulmate - Tuesday, March 24, 2015 - link

    PFM is so much needed for ocz drives... almost all my dead vector drives are because of sudden power loss... Reply
  • ocztosh - Tuesday, March 24, 2015 - link

    As this is my first post on this article please allow me to identify myself as an OCZ representative.

    Hi Zodiacsoulmate thank you for your comment. We do feel like PFM+ helps add more protection for power users and ultimately it was our goal with this drive to create a product that delivered both high performance and enhanced reliability by leveraging some of the features from our enterprise drives.

    Thank you for your business and we are sorry that you have had issues with any of our drives in the past and in regards to the very first Vector drives we have released a firmware update that addresses sudden power loss issues. If you currently are experiencing any issues please do not hesitate to contact our support team and we will be happy to support you.

    Over the last two years we have made significant upgrades in our products and processes to impact quality and continue to make investments in this area. All of our drives leverage premium Toshiba NAND and that complete access to next generation flash is enabling us to develop some very exciting new solutions that are both focused on enhancing performance and driving down cost.
    Reply
  • zodiacsoulmate - Tuesday, March 24, 2015 - link

    wow cool man, yea quite some vectors died but the 5 year warranty got them all replace, one 256gb died so many times that OCZ replaced it with a vector150 which is running strong since then... Reply
  • ocztosh - Tuesday, March 24, 2015 - link

    Hi Zodiacsoultmate. Thanks very much for the feedback and glad to hear that your Vector 150 has been running great. Again sorry that you had issues with the original Vector and thank you very much for your business. We have made significant changes and improvements to both products and processes and will continue to moving forward. Thanks for sticking with us. Reply
  • Spoogie - Tuesday, March 24, 2015 - link

    Glad to see you're watching.

    I've had two failed OCZ drives, one DOA and the replacement 8 months later. The packaged replacement (unopened) was sold on eBay out of fear, at which point OCZ was sworn off.

    I believe in redemption, and I'll be watching the consumer feedback. Good luck.
    Reply
  • ocztosh - Tuesday, March 24, 2015 - link

    Hi Spoogie. Thank you for your comments and your business. I am sorry to hear that you had issues in the past. I know it is easy to say but we are a completely different company/organization today as part of the Toshiba Group. We certainly realize we need to continue to work diligently to support our customers and appreciate the feedback and open dialogue. Reply
  • Oxford Guy - Wednesday, March 25, 2015 - link

    I had two expensive (at the time) Vertex 2 drives. One bricked after about two weeks. The RMA drive bricked immediately when updating the firmware. I have the replacement drive but I don't trust it enough to use it. The other drive I bought worked through several firmware updates (a few months) but bricked after I put my second Macbook Pro to sleep. I was busy with graduate school and a job and didn't have a chance to update my backup and part of the next day's final presentation for a course and also some important poetry. I had been lulled into a false sense of security because the drive had operated with no trouble for several months.

    When Anandtech wrote about how OCZ started selling Vertex 2 drives with half the NAND chips in them, without changing the specs, it didn't say anything about reduced reliability. However, from my experience and from what I've seen from comments on the Net (such as Amazon reviews), the failure rate for these drives was extremely high. Also, when OCZ offered a replacement option for those burned by the bait and switch hidden 32-bit NAND (full chip complement) to 64-bit NAND (half the chips) transition -- it never offered it for the 240 GB model. One of Anandtech's writers also said they didn't experience failures with their Vertex 2 drives but those were most likely all 32-bit NAND drives.

    OCZ never released a utility for dealing with bricked Sandforce drives due to their "panic mode" that would help users get some data recovered. Instead, their forum staff just said "RMA or send to DriveSavers". From what I gathered from my Google searches, the panic mode (which Anandtech's Vertex 2 reviews didn't mention as I recall) was designed to protect Sandforce's flaky firmware from reverse-engineering at the cost of making it extremely difficult (impossible for consumers without spending thousands for DriveSavers) to obtain their data. And, since these drives love to corrupt themselves at the drop of a hat (possibly due to deduplication of critical filesystem data), it's unlikely that most of the panicking is caused by serious problems with the drives like the actual failure of a NAND chip.

    So, it seems like I received a product that not only had a poorly-engineered anti-consumer controller but which also had half the NAND chips and thus the loss of performance, capacity, and apparently also stability. I think these Vertex drives were rated at 1 1/2 stars on Amazon or something similarly low. I also remember reading about a lot of problems with Vertex 3/Agility 3 drives so it seems that the continuing defectiveness of the OCZ/Sandforce combination moved forward past the Vertex 2 era and the first-generation controller.
    Reply
  • Oxford Guy - Wednesday, March 25, 2015 - link

    "it's unlikely that most of the panicking is caused by serious problems with the drives like the actual failure of a NAND chip."

    Although there is also a rumor that part of the failure rate is due to OCZ using its own NAND which may not be as reliable as less generic NAND. Rumors abound but concrete information is hard to come by because the whole mess was brushed under the rug.
    Reply
  • Oxford Guy - Wednesday, March 25, 2015 - link

    I was also the lucky owner of an IBM 75GXP "Deathstar" hard disk so I have an almost supernatural knack for picking the worst possible drives for my data apparently. Like the Vertex 2, the Deathstar was hyped by reviews for being so wonderfully quick — until they started to blow up. Reply
  • AssBall - Wednesday, March 25, 2015 - link

    I had 2 Deathstars and both worked for 4 years. They probably still work. Guess I was lucky. Reply

Log in

Don't have an account? Sign up now