The SandForce Roundup: Corsair, Kingston, Patriot, OCZ, OWC & MemoRight SSDs Compared
by Anand Lal Shimpi on August 11, 2011 12:01 AM ESTAnandTech Storage Bench 2011 - Light Workload
Our new light workload actually has more write operations than read operations. The split is as follows: 372,630 reads and 459,709 writes. The relatively close read/write ratio does better mimic a typical light workload (although even lighter workloads would be far more read centric).
The I/O breakdown is similar to the heavy workload at small IOs, however you'll notice that there are far fewer large IO transfers:
| AnandTech Storage Bench 2011 - Light Workload IO Breakdown | ||||
| IO Size | % of Total | |||
| 4KB | 27% | |||
| 16KB | 8% | |||
| 32KB | 6% | |||
| 64KB | 5% | |||
Despite the reduction in large IOs, over 60% of all operations are perfectly sequential. Average queue depth is a lighter 2.2029 IOs.
As the only 240GB drive running 3.20 the Kingston HyperX is the fastest drive in our test here (only the Vertex 3 and possibly the Corsair Force GT even have 3.20 available for update). The advantage is still pretty small though, at 6% above its closest competitor in an IO bound test you'll never see that advantage in the real world.
Facebook
Twitter
RSS
90 Comments
View All Comments
bobbyh - Thursday, August 11, 2011 - link
FIRST!Are you going to talk about synchronous vs asynchronous NAND and the benefits of one vs the other?
bobbyh - Thursday, August 11, 2011 - link
nevermind lol!bobbyh - Thursday, August 11, 2011 - link
very nice roundup A+ would read againArnulf - Thursday, August 11, 2011 - link
FIRST what ? FIRST idi0t to tag himself ? You got that right !ARoyalF - Thursday, August 11, 2011 - link
The estimated cost breakdown sure gave me an appreciation of what goe$ on behind the scenes.Sagath - Thursday, August 11, 2011 - link
Firstly, I'd state I always appreciated you bringing these issues to the front page to allow the consumer to see these issues in a public venue, while also berating manufacturers for selling us junk. Thank you, Anand.That being said; I fully understand that the new Sandforce chips allow SATA6 connectivity, and are thus the fastest possible drives on the market...yet I have to ask, is it worth it? I don't see you mentioning these issues with last gens drives like the aforementioned X25-m, or Sandforce v1.
Any SSD sold today is plainly 'fast', and order of magnitudes faster then magnetic-based storage. Is the incremental upgrade (of microseconds at best?) really worth sacrificing the reliability associated with last generations drives?
My X25-M and Vertex 2's across multiple computers, laptops and friends computers are all running flawlessly. I have had zero complaints about random BSOD's or lockups. I also have 2 friends with whom purchased Vertex 3's on their own, and are both experiencing the famous Sandforce v2 issues...
I'll stick with my 'slower' (lol?) X25-m's and V2's, then deal with these issues.
bobbyh - Thursday, August 11, 2011 - link
I have an older x25-m it still works flawlessly, this generation of drives has had an insane amount of problems.tbanger - Thursday, August 11, 2011 - link
Can anyone shed some more light on the Intel 320 series firmware problem that Anand mentions?I've experienced it recently myself with my work machine's 300GB model resetting itself to an 8MB partition with all data lost. Not a huge problem (good backup scheme) but still annoying. At least Intel kindly replaced my drive with a new one fairly quickly. However, given I had already ordered a bunch more drives for the company (before the failure), I would like to see a firmware update that fixes this problem. I'm getting nervous that we're going to experience a bunch of failures.
Is there any official plan to fix this from Intel? I haven't found much from Googling other than user complaints with little response from Intel.
Nickel020 - Thursday, August 11, 2011 - link
Just follow the link in the article ;)http://communities.intel.com/message/133499
They've reproduced the issue and are validating the firmware fix. I got not clue how long their validating could take, but a new FW could be out any day, or maybe it'll take another month. They might find some issues during validation, which need further fixes and then further validating, so not even someone from Intel could give you a definite ETA.
tbanger - Thursday, August 11, 2011 - link
That'll teach me to only skim the article :)Thanks for the link. Nice to see Intel to offer a little official feedback.