Impact of SSD Size on Cache Performance

A cache's performance is determined by how well the caching algorithm matches the workload, how well the cache size accommodates the workload and how quickly you can access the cache. So far we've demonstrated that Corsair's F40 can perform anywhere from noticeably worse to a bit faster than Intel's SSD 311 as a cache. I wanted to see if I could demonstrate the capacity advantage alone offered by the F40. To illustrate this I performed 11 tasks in a row composed of application launches, game launches and level loads. After running through all 11 tasks once, I rebooted and ran through them again. I then rebooted again and ran through the 11 tasks one more time, for a total of two cached runs per SSD cache. In theory the larger cache should be able to maintain application launch performance between runs:

Application Launch Performance - Seagate Barracuda 3TB - Time in Seconds
Application (Listed in Launch Order) First Run Intel SSD 311 - Run 1 Intel SSD 311 - Run 2 Corsair Force F40 - Run 1 Corsair Force F40 - Run 2
Portal 2 12.0 9.6 12.1 9.3 9.3
Portal 2 (Level Load) 17.1 14.8 15.4 14.1 14.2
WoW 5.3 4.3 5.8 4.7 4.4
WoW (Level Load) 11.9 11.7 6.2 5.2 5.2
Starcraft 2 15.3 11.9 10.2 7.9 8.0
Starcraft 2 (Level Load) 23.3 17.6 14.1 14.4 15.1
Photoshop CS5.5 7.1 3.5 6.5 6.3 3.2
After Effects CS5.5 19.3 6.6 11.6 12.7 6.7
Dreamweaver CS5.5 8.0 4.7 6.2 6.2 4.5
Illustrator CS5.5 6.1 2.9 3.7 5.3 3.1
Premier Pro CS5.5 10.4 3.2 3.1 6.9 3.2

Application Launch Test - Maximized Cache

On the first run both the Corsair F40 and Intel SSD 311 manage to perform very similarly. On the second run however, many of the applications load slower on the SSD 311 as some data has apparently been evicted from the cache. The 40GB Corsair F40 doesn't suffer the same fate as it has much more data for storage and as a result the second run through shows continued performance improvement.

AnandTech Storage Bench 2011 - Light Workload Final Words
POST A COMMENT

81 Comments

View All Comments

  • jordanclock - Friday, May 13, 2011 - link

    I like the idea of this article and it starts providing some extra data that was asked for in the original review: What about other SSDs? Could we get some more SSDs tested with SRT? I'm not expecting Vertex 3s, but some "older" SSDs like the F40 that would likely be replaced soon. Reply
  • MrSpadge - Friday, May 13, 2011 - link

    Currently Anand tested 3 mid- to high performance SSDs and I think this gives us a pretty good picture overall. However, if another one was tested I'd want the 64 GB Agility 3 and/or Solid 3. Well, a proper review of them would be very nice anyway ;)

    MrS
    Reply
  • Cow86 - Friday, May 13, 2011 - link

    Yea I was thinking the same thing...they supposedly perform a lot better, while still being reasonably affordable for the 60GB drives (~110 euro's here). Maybe they'd be the ultimate cache at a reasonable price point? Either way would love a review of those as well, vertex 3 is staying rather expensive :( Reply
  • jebo - Friday, May 13, 2011 - link

    I agree that the Agility 3 and Solid 3 look very promising. Reply
  • therealnickdanger - Friday, May 13, 2011 - link

    Plus, SRT works with drives up to 64GB, so it seems worth it to test a drive of that size. I imagine there are many users (such as myself) that have older Intel 80GB, Indilix 64GB, or SandForce SF1200/1500 64GB drives that will soon be replaced with faster 6Gbps drives.

    I would imagine that you could a drive of any size, especially the larger ~120GB SSDs that offer significant speed advantages over the 64GB models. SRT would limit the usable cache to 64GB, but it would still be interesting to see...

    *wink**hint*
    Reply
  • Boissez - Friday, May 13, 2011 - link

    Yup - Here's another vote for that article. I have an 'old' 60GB agility 2 boot drive and a Z68 board on it's way. Question now is whether I keep it as boot drive or whether I take a performance hit and use it as cache. This article comes close to answering it yet Anandtech's own SSD workload benchmarks (which are the most interesting IMO) does not include the F40's numbers. Reply
  • GullLars - Tuesday, May 24, 2011 - link

    If you can fit your OS and core apps/games on your agility 2, it's a no-brainer, keep it as you have. If you have a lot of apps and games you have had on HDDs and use often, you could concider trying it as cache.
    I have 3 computers with 32GB boot drives (1 laptop, HTPC, and a donated SSD to my fathers computer), and all of them get by fine with that with a little bit of management. You could concider using 30GB for caching and 30GB for OS + core apps, getting full SSD speed on the stuff you use the most, and have cache help out on game loads and more seldom used apps.
    Reply
  • Mr Perfect - Friday, May 13, 2011 - link

    That's what I'm thinking too. Get yourself the largest, fastest cache you can and see what happens. :) Currently that's looking like a 64GB Agility 3 or Solid 3. Reply
  • samsp99 - Friday, May 13, 2011 - link

    I was just asking myself this very question... Reply
  • genji256 - Friday, May 13, 2011 - link

    I'm confused. If the 311 performed better on the second run ("Intel 311 - Run 1" in your table), shouldn't that mean that the cache was large enough to store the data for all the applications (since they all ran during the first run)? If so, why would there suddenly be data evicted from the cache between the second and third run ("Intel 311 - Run 1" and "Intel 311 - Run 2" respectively)? Reply

Log in

Don't have an account? Sign up now