AnandTech Storage Bench 2011

I didn't expect to have to debut this so soon, but I've been working on updated benchmarks for 2011. Last year we introduced our AnandTech Storage Bench, a suite of benchmarks that took traces of real OS/application usage and played them back in a repeatable manner. I assembled the traces myself out of frustration with the majority of what we have today in terms of SSD benchmarks.

Although the AnandTech Storage Bench tests did a good job of characterizing SSD performance, they weren't stressful enough. All of the tests performed less than 10GB of reads/writes and typically involved only 4GB of writes specifically. That's not even enough exceed the spare area on most SSDs. Most canned SSD benchmarks don't even come close to writing a single gigabyte of data, but that doesn't mean that simply writing 4GB is acceptable.

Originally I kept the benchmarks short enough that they wouldn't be a burden to run (~30 minutes) but long enough that they were representative of what a power user might do with their system.

Not too long ago I tweeted that I had created what I referred to as the Mother of All SSD Benchmarks (MOASB). Rather than only writing 4GB of data to the drive, this benchmark writes 106.32GB. It's the load you'd put on a drive after nearly two weeks of constant usage. And it takes a *long* time to run.

I'll be sharing the full details of the benchmark in some upcoming SSD articles but here are some details:

1) The MOASB, officially called AnandTech Storage Bench 2011 - Heavy Workload, mainly focuses on the times when your I/O activity is the highest. There is a lot of downloading and application installing that happens during the course of this test. My thinking was that it's during application installs, file copies, downloading and multitasking with all of this that you can really notice performance differences between drives.

2) I tried to cover as many bases as possible with the software I incorporated into this test. There's a lot of photo editing in Photoshop, HTML editing in Dreamweaver, web browsing, game playing/level loading (Starcraft II & WoW are both a part of the test) as well as general use stuff (application installing, virus scanning). I included a large amount of email downloading, document creation and editing as well. To top it all off I even use Visual Studio 2008 to build Chromium during the test.

Update: As promised, some more details about our Heavy Workload for 2011.

The test has 2,168,893 read operations and 1,783,447 write operations. The IO breakdown is as follows:

AnandTech Storage Bench 2011 - Heavy Workload IO Breakdown
IO Size % of Total
4KB 28%
16KB 10%
32KB 10%
64KB 4%

Only 42% of all operations are sequential, the rest range from pseudo to fully random (with most falling in the pseudo-random category). Average queue depth is 4.625 IOs, with 59% of operations taking place in an IO queue of 1.

Many of you have asked for a better way to really characterize performance. Simply looking at IOPS doesn't really say much. As a result I'm going to be presenting Storage Bench 2011 data in a slightly different way. We'll have performance represented as Average MB/s, with higher numbers being better. At the same time I'll be reporting how long the SSD was busy while running this test. These disk busy graphs will show you exactly how much time was shaved off by using a faster drive vs. a slower one during the course of this test. Finally, I will also break out performance into reads, writes and combined. The reason I do this is to help balance out the fact that this test is unusually write intensive, which can often hide the benefits of a drive with good read performance.

There's also a new light workload for 2011. This is a far more reasonable, typical every day use case benchmark. Lots of web browsing, photo editing (but with a greater focus on photo consumption), video playback as well as some application installs and gaming. This test isn't nearly as write intensive as the MOASB but it's still multiple times more write intensive than what we were running last year.

As always I don't believe that these two benchmarks alone are enough to characterize the performance of a drive, but hopefully along with the rest of our tests they will help provide a better idea.

The testbed for Storage Bench 2011 has changed as well. We're now using a Sandy Bridge platform with full 6Gbps support for these tests. All of the older tests are still run on our X58 platform.

AnandTech Storage Bench 2011 - Heavy Workload

We'll start out by looking at average data rate throughout our new heavy workload test:

AnandTech Storage Bench 2011 - Heavy Workload

In our heavy test for 2011 the 120GB Vertex 3 is noticeably slower than the 240GB sample we tested a couple of months ago. Fewer available die are the primary explanation. We're still waiting on samples of the 120GB Intel SSD 320 and the Crucial m4 but it's looking like this round will be more competitive than we originally thought.

The breakdown of reads vs. writes tells us more of what's going on:

AnandTech Storage Bench 2011 - Heavy Workload

Surprisingly enough it's not read speed that holds the 120GB Vertex 3 back, it's ultimately the lower (incompressible) write speed:

AnandTech Storage Bench 2011 - Heavy Workload

The next three charts just represent the same data, but in a different manner. Instead of looking at average data rate, we're looking at how long the disk was busy for during this entire test. Note that disk busy time excludes any and all idles, this is just how long the SSD was busy doing something:

AnandTech Storage Bench 2011 - Heavy Workload

AnandTech Storage Bench 2011 - Heavy Workload

AnandTech Storage Bench 2011 - Heavy Workload

Random & Sequential Performance AnandTech Storage Bench 2011 - Light Workload


View All Comments

  • Xcellere - Wednesday, April 06, 2011 - link

    It's too bad the lower capacity drives aren't performing as well as the 240 GB version. I don't have a need for a single high capacity drive so the expenditure in added space is unnecessary for me. Oh well, that's what you get for wanting bleeding-edge tech all the time. Reply
  • Kepe - Wednesday, April 06, 2011 - link

    If I've understood correctly, they're using 1/2 of the NAND devices to cut drive capacity from 240 GB to 120 GB.
    My question is: why don't they use the same amount of NAND devices with 1/2 the capacity instead? Again, if I have understood correctly, that way the performance would be identical compared to the higher capacity model.
    Is NAND produced in only one capacity packages or is there some other reason not to use NAND devices of differing capacities?
  • dagamer34 - Wednesday, April 06, 2011 - link

    Because price scaling makes it more cost-effective to use fewer, more dense chips than separate smaller, less dense chips as the more chips made, the cheaper they eventually become.

    Like Anand said, this is why you can't just as for a 90nm CPU today, it's just too old and not worth making anymore. This is also why older memory gets more expensive when it's not massively produced anymore.
  • Kepe - Wednesday, April 06, 2011 - link

    But couldn't they just make smaller dies? Just like there are different sized CPU/GPU dies for different amounts of performance. Cut the die size in half, fit 2x the dies per wafer, sell for 50% less per die than the large dies (i.e. get the same amount of money per wafer). Reply
  • A5 - Wednesday, April 06, 2011 - link

    No reason for IMFT to make smaller dies - they sell all of the large dies coming out of the fab (whether to themselves or 3rd parties), so why bother making a smaller one? Reply
  • vol7ron - Wednesday, April 06, 2011 - link

    You're missing the point on economies of scale.

    Having one size means you don't have leftover parts, or have to pay for a completely different process (which includes quality control).

    These things are already expensive, adding the logistical complexity would only drive the prices up. Especially, since there are noticeable difference in the manufacturing process.

    I guess they could take the poorer performing silicon and re-market them. Like how Anand mentioned that they take poorer performning GPUs and just sell them at a lower clockrate/memory capacity, but it could be that the NAND production is more refined and doesn't have that large of a difference.

    Regardless, I think you mentioned the big point: inner RAIDs improve performance. Why 8 chips, why not more? Perhaps heat has something to do with it, and (of course) power would be the other reason, but it would be nice to see higher performing, more power-hungry SSDs. There may also be a performance benefit in larger chips too, though, sort of like DRAM where 1x2GB may perform better than 2x1GB (not interlaced).

    I'm still waiting for the manufacturers to get fancy, perhaps with multiple controllers and speedier DRAM. Where's the Vertex3 Colossus.
  • marraco - Tuesday, April 12, 2011 - link

    Smaller dies would improve yields, and since they could enable full speed, it would be more competitive.

    A bigger chip with a flaw may invalidate the die, but if divided in two smaller chips it would recover part of it.

    On other side, probably yields are not as big problem, since bad sectors can be replaced with good ones by the controller.
  • dfhjtykyu - Tuesday, April 19, 2011 - link

    hello, Welcome to the online store ^__^:

    ^__^:====( www etradinglife com )======

    1) More pictures available on our website =
    3) Perfect quality, small order accepted .
    4) 100% safe door to door delivery, within 5 - 7 days air express for small orders .

    5) We have lots of jerseys in stock

    6) Letters and number are sewn on b2cshop body, 100% embroidery

    7) Size: .48, 50, 52, 54, 56, 60

    8) Delivery by UPS, DHL, EMS door to door

    9) Delivery in 5 - 7 days

    NFL,NBA,MLB all are 18usd!!!!
  • Kepe - Wednesday, April 06, 2011 - link

    Anand, I'd like to thank you on behalf of pretty much every single person on the planet. You're doing an amazing job with making companies actually care about their customers and do what is right.
    Thank you so much, and keep up the amazing work.

    - Kepe
  • dustofnations - Wednesday, April 06, 2011 - link

    Thank God for a consumer advocate with enough clout for someone important to listen to them.

    All too often valid and important complaints fall at the first hurdle due to dumb PR/CS people who filter out useful information. Maybe this is because they assume their customers are idiots, or that it is too much hassle, or perhaps don't have the requisite technical knowledge to act sensibly upon complex complaints.

Log in

Don't have an account? Sign up now