Random Read/Write Speed

The four corners of SSD performance are as follows: random read, random write, sequential read and sequential write speed. Random accesses are generally small in size, while sequential accesses tend to be larger and thus we have the four Iometer tests we use in all of our reviews.

Our first test writes 4KB in a completely random pattern over an 8GB space of the drive to simulate the sort of random access that you'd see on an OS drive (even this is more stressful than a normal desktop user would see). I perform three concurrent IOs and run the test for 3 minutes. The results reported are in average MB/s over the entire time. We use both standard pseudo randomly generated data for each write as well as fully random data to show you both the maximum and minimum performance offered by SandForce based drives in these tests. The average performance of SF drives will likely be somewhere in between the two values for each drive you see in the graphs. For an understanding of why this matters, read our original SandForce article.

Iometer - 4KB Random Write, 8GB LBA Space, QD=3

Peak performance on the 120GB Vertex 3 is just as impressive as the 240GB pre-production sample as well as the m4 we just tested. Write incompressible data and you'll see the downside to having fewer active die, the 120GB drive now delivers 84% of the performance of the 240GB drive. In 3Gbps mode the 240 and 120GB drives are identical.

Many of you have asked for random write performance at higher queue depths. What I have below is our 4KB random write test performed at a queue depth of 32 instead of 3. While the vast majority of desktop usage models experience queue depths of 0 - 5, higher depths are possible in heavy I/O (and multi-user) workloads:

Iometer - 4KB Random Write, 8GB LBA Space, QD=32

At high queue depths the gap between the 120 and 240GB Vertex 3s grows a little bit when we're looking at incompressible data.

Iometer - 4KB Random Read, QD=3

Random read performance is what suffered the most with the transition from 240GB to 120GB. The 120GB Vertex 3 is slower than the 120GB Corsair Force F120 (SF-1200, similar to the Vertex 2) in our random read test. The Vertex 3 is actually about the same speed as the old Indilinx based Nova V128 here. I'm curious to see how this plays out in our real world tests.

Sequential Read/Write Speed

To measure sequential performance I ran a 1 minute long 128KB sequential test over the entire span of the drive at a queue depth of 1. The results reported are in average MB/s over the entire test length.

Iometer - 128KB Sequential Write

Highly compressible sequential write speed is identical to the 240GB drive, but use incompressible data and the picture changes dramatically. The 120GB has far fewer NAND die to write to in parallel and in this case manages 76% of the performance of the 240GB drive.

Iometer - 128KB Sequential Read

Sequential read speed is also lower than the 240GB drive. Compared to the SF-1200 drives there's still a big improvement as long as you've got a 6Gbps controller.

The Vertex 3 120GB AnandTech Storage Bench 2011


View All Comments

  • Kepe - Wednesday, April 06, 2011 - link

    I'd say the reason is usually that when a company has sold you its product, they suddenly lose all interest in you until they come up with a new product to sell. Apple used to be a very good example with its battery policy. "So, your battery died? We don't sell new or replace dead batteries, but you can always buy the new, better iPod."
    It's this kind of ignorance towards the consumers that is absolutely appalling, and Anand is doing a great job at fighting for the consumer's rights. He should get some sort of an award for all he has done.
  • Super - Friday, April 08, 2011 - link

    ...perhaps the Nobel Peace Prize. ?? i've seen someone win it for a whole lot less *cough Obama Reply
  • A5 - Wednesday, April 06, 2011 - link

    Agreed - glad they listen to Anand.

    The real question is why they didn't do anything until Anand bitched to the CEO directly. It's not like they weren't aware of the issue - the Storage Review article came out several months ago...
  • darckhart - Wednesday, April 06, 2011 - link

    It just goes to show that companies are not customer focused. Unless they get shoved hard enough, or see that the bottom line will be affected greatly, they just hope you'll give up after being mired in the revolving email chain or sent through 5 level deep phone support.

    Thanks Anand for reminding companies that some of us are still capable of making informed decisions and aren't afraid to express our dissatisfaction with our dollars.
  • 789427 - Thursday, April 07, 2011 - link

    It's not about being customer focussed or not. Quite frankly, what percentage of upgraders will go into this level of detail?

    Furthermore, 25nm sounds better than 35nm to most people and that's salesmen included.

    After all that, it's a victory for transparency for a tiny few.

    In terms of marketing, there's little you can do except re-brand the entire product range.

    e.g. Silver and Silver Pro for the lower capacities, Gold and Gold Pro for the higher capacities and explain on the box that fewer chips means generally lower performance

    The problem here is that this is the cutting edge of technology and that in 12 months time, it will be surpassed. Then how do you re-vamp the line?

    Graphics cards have this problem too and the model numbers are baffling for 99% of first-time buyers.

    What I would advocate is a sticker valid for 3 months on the product that lets you know which product in terms of performance you are buying and a URL you can visit to check for an update.


    Your product: xyz 300-35
    is better than xyz 300-24
    but is worse than 300-ii

    Check Real performance figures here: URL

    Then it would be nice for salesmen to allow customers to verify this.
  • cactusdog - Wednesday, April 06, 2011 - link

    Yep, at least OCZ have made a commitment not to use slow hynix nand and are being more transparent with real world performance but its all too little too late.

    Branding drives with the 25 or 34nm prefix is redundant now that all(or most) nand being produced is 25nm. Ocz made no real attempt to fix the problem when they needed to, and continued to sell the drives even after the consumer backlash.

    I disagree with Anand that other manufacturers of sandforce controller drives hide the specs as OCZ did. Corsair rebranded their 25nm drives from the start. Other non sandforce drives from Intel also rebranded their 25nm drives.

    Its true that many companies use different components and use the same branding but rarely does the performance vary as much as 30%. 30% is a huge and not acceptable for high end expensive parts..

    Its a pity Anand didnt really have anything to add on the Spectek issue that hasnt already been said. I find it hard to believe a company like Micron would sell very expensive nand cheaper to Spectek unless there is some problem with it.

    Saying Spectek nand must be OK because it is still rated at 3000 cycles doesnt sound very thorough or tell us the whole story. The cycle rating could have very different testing standards between Micron and Spectek.

    I would have thought it would be easy for someone like Anand to ask Micron or Spectek if the Spectek nand is tier 1 nand or not. I wouldnt trust OCZ response given their track record.

    Overall though thanks Anand for sticking up for the consumers.
  • Powerlurker - Wednesday, April 06, 2011 - link

    According to their corporate website "SpecTek began at Micron in 1988 as a component-recovery group." which would lead me to believe that they're Micron's low-end brand for disposing of lower performing dies. Reply
  • Xneb - Thursday, April 07, 2011 - link

    That is correct. testing is the same though so end users should not be able to tell the difference between spectek and imft nand in these drives. Reply
  • sleepeeg3 - Thursday, April 07, 2011 - link

    You can't fault him for reporting honestly. There is no concrete data that shows Spectek NAND is inferior to Micron. Reply
  • Alkapwn - Wednesday, April 06, 2011 - link

    Ditto! Keep up the great work! We all appreciate it greatly! Reply

Log in

Don't have an account? Sign up now