Intel SSD 520 Review: Cherryville Brings Reliability to SandForce
by Anand Lal Shimpi on February 6, 2012 11:00 AM ESTPerformance Over Time & TRIM
SandForce's controllers have always behaved poorly if you pushed them into their worst case scenario. Should you fill a SF-2281 based drive completely with incompressible data then continue to write to the drive with more incompressible data (overwriting some of what you've already written) to fill up the spare area you'll back the controller into a corner that it can't get out of, even with TRIM. It's not a terribly realistic situation since anyone using an SF-2281 SSD as a boot drive will at least have Windows (or some other easily compressible OS) installed, and it's fairly likely that you'll have other things stored on your SSD in addition to large movies/photos. Regardless, it's a corner case that we do have to pay attention to.
I was curious to see if Intel's firmware did anything differently than the standard SandForce build used by other partners. To find out I took an Intel SSD 520 (240GB) and a Kingston HyperX (SF-2281 240GB) and filled both drives with incompressible data. I then ran a 60 minute 4KB random write torture test (QD32), once again, with incompressible data. Normally I'd use HDTach to chart performance over time however HDTach measures performance with highly compressible data so we wouldn't get a good representation of performance here. Instead I ran AS-SSD to get a good idea for incompressible sequential performance in this worst case state. Afterwards, I TRIMed the drives and ran AS-SSD again to see if TRIM could recover the drive's performance.
| Intel SSD 520 - Resiliency - AS SSD Sequential Write Speed - 6Gbps | |||||
| Clean | After Torture | After TRIM | |||
| Intel SSD 520 240GB | 284.5 MB/s | 162.9 MB/s | 162.9 MB/s | ||
| Kingston HyperX 240GB | 289.3 MB/s | 162.5 MB/s | 162.5 MB/s | ||
Surprisingly enough, Cherryville doesn't actually perform any differently than the stock SandForce firmware in this case. SandForce definitely improved the worst case scenario performance with the SF-2281 compared to the first generation controller, and it also improved performance compared to earlier builds of the 2281 firmware, but Intel appears to have not done anything above and beyond here. Based on these results I'd be willing to bet that Intel doesn't have source code access to the SF-2281 firmware otherwise it would've worked on a solution to this corner case. Performance in this worst case scenario isn't terrible but the fact that it's irrecoverable even after a TRIM is what's most troubling. Again, I don't see most end users backing themselves into this corner but it's worth pointing out.
Facebook
Twitter
RSS
138 Comments
View All Comments
ckryan - Tuesday, February 7, 2012 - link
True, but the 520 series isn't the 320 series. Intel has the 510 and 520 series to satisfy the user that desires speed, but the fact is Intel will sell more SATA II than SATA III drives for the next few years.lyeoh - Tuesday, February 7, 2012 - link
Maybe making fast expensive drives and selling them for profit isn't Intel's main goal. Perhaps Intel is trying to push the SSD market to the next step.Nowadays drives are often the biggest performance bottleneck, and so it stops making a big difference whether you buy a top of the line Intel CPU or not, or even buy AMD instead.
But once fast SSDs start becoming more reliable and sell at closer to _commodity_ prices, then Intel can get back to selling powerful CPUs to more people.
A marketplace full of fast expensive SSDs that BSOD and lose data and slow cheap hard drives doesn't help for such a goal.
But helping Sandforce _eventually_ get their act together (Intel aren't doing this for free of course) might do so. They might also be concerned about Samsung.
NitroWare - Wednesday, February 8, 2012 - link
"Perhaps Intel is trying to push the SSD market to the next step."Intel has frequently done this with different technologies.So it is logical.
In the article, Intel claimed they had 100% of staff on Intel SSD. Vendors usually do not make claims about internal employee hardware.
Ethernet, Wifi, PCI, Legacy free, non x86 ISA (Xscale, Itanic), Fibre, PC I/O and RAM Technologies.
They even dabbled with video conferencing years ago
gramboh - Monday, February 6, 2012 - link
I'm in the market for a 240gb SSD in the next few months, and would love to buy a 520 series over other SF-2281/Marvell options, but right now it's a 42% price premium which is way too much.The 120gb has a more realistic price premium of about 25% (taking into account rebates on the 2281/Marvell drives) which I would consider, would have been nice for Intel to price the 240gb at $400-430.
mckirkus - Wednesday, February 29, 2012 - link
"Would have been nice for Intel to price the 240gb at $400-430"The 240gb version is $357 on Amazon right now (2/29).
LeftSide - Monday, February 6, 2012 - link
I bought the Kingston version on Black Friday for $120 after MIR. It was a great deal. Although I did worry about the BSOD (especially right after sending in the MIR). Thankfully, I haven't had any problems and it has been an excellent drive. It's unbelievably fast.NitroWare - Wednesday, February 8, 2012 - link
They have the 332 firmware available for download from their website. There is even a toolbox utility now but this only has SMART features and no wipe tool.jaydee - Monday, February 6, 2012 - link
Would you expect the 120GB version to perform similar to the 60GB or the 240GB version? Or right down the middle?extide - Monday, February 6, 2012 - link
Historically, 128GB drives have seemed to perform closer to the 240/256GB counterparts than the 60/64GB ones. That's just my observation, though.ckryan - Monday, February 6, 2012 - link
It's closer to the 240GB in performance. The 60GB SF drives are really at a disadvantage, especially in sequential writes. There have been benches of the 120GB 520 for some time, but the 240GB is unquestionable faster. You might not notice the difference in real usage. You can look at Anand's 120GB roundup from last year, and the 120GB version should come out slightly ahead of the Vertex 3.