OCZ Vertex 3 MAX IOPS & Patriot Wildfire SSDs Reviewed
by Anand Lal Shimpi on June 23, 2011 4:35 AM ESTAnandTech Storage Bench 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.
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.
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:
The breakdown of reads vs. writes tells us more of what's going on:
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:
Facebook
Twitter
RSS
112 Comments
View All Comments
lyeoh - Thursday, June 23, 2011 - link
Because the main reasons why so very many SSD drives fail does not appear to be due to the "nm" of the NANDs.As far as I know, the larger "nm" SSDs also have been failing a lot.
Movieman420 - Thursday, June 23, 2011 - link
Can't wait until you guys get your hands on the hard to find/unavailable 240gb MaxIOP drive. It will obviously be the king once it hits the channel. Farther down the road, I dare say that the MaxIOP will be bested. How? you ask? The answer...the soon to be available Samsung 27nm DDR toggle nand said to out run the 32nm SDR toggle by 40-60 percent as well as match the durability (p/e cycles) of 34nm nand! Now the big question is...Does the SF2281 have enuff uuumph left to utilize the much faster DDR nand? I'm hoping that the lower cost of the 27nm process nand will offset the 'fastest nand' price premium that will surely be levied. Also, the new 27nm DDR will be available in both 32Gb and 64Gb dies which will allow there to be a higher performance 60GB SF2281 drive...atm the only Ocz 60gb SF2281s are the Agility 3 and Solid 3 which are almost beaten by the Vertex 2 34nm equivalent...that's why they didn't earn the Vertex 3 name tag. A higher performing small drive is just the thing that raid fanatics like me like to use. You can have a fast array without spending $300-$500 per drive for real sata III performance. This is why a 60gb Vertex 3 drive would sell like wild fire. Don't get me wrong, my current 4 x 34nm Vertex 2 60gb array is plenty fast but don't have near the compressed write resilience (aka less nand throttling) of the SF2281 drives. I look forward to spending ~$300 for 2 V3 60's that will totally blow away a single $300 120gb/$500 240gb V3.L. - Thursday, June 23, 2011 - link
I don't quite get what you are trying to say.But basically, DDR = RAID0 2 elements .
So between 8 dies on an 8 channel controller and 8 ddr dies on a 4 channel controller you should see absolutely no difference.
Now considering current controllers have 8+ channels and the smallest ddr at the expected time of implementation is a dual die 32gb nand chip .. that's 240GB and abbove.
This is not the world of RAM where you can just stick more chips in parallel (ddr5...), then dual channel controller over that etc.
Of course the same ideas are always applicable, but for those samsung ddr chips to be worth anything you'd need to go to 480GB+ and no they're not faster than anything they're just ddr, and that difference matters a lot.
You will not find two 60GB drives that are better in raid0 than a single 120GB drive, because said 120GB drive is basically those two drives in raid0, with one less controller and the raid on the sandforce controller, better algorithm usage etc.
Again, number of channels, nand packages, etc.
Unless somebody starts making 32 GB DDR5 NAND dies I don't expect much change in there ;)
Gittun - Thursday, June 23, 2011 - link
You do realize the 240GB max iops edition is already in the graphs?The_True - Thursday, June 23, 2011 - link
Well!!i wont say anything bad about anand, but i think anad is putting his hand on fire for OCZ SSD.
i used to like OCZ, but i get tired of RMA and windows Reinstall every other week.
i dont care about 20 or 30 MB/S faster, i care about relaibility, not benchmark the driver every hour to make me feel good.
OCZ driver are not reliable, if you use your computer for work, go Intel is the only SSD out there that is reliable for business.
MAX IOPS?? i called it MAX windows re install!!
after getting two intel 510 250Gb, i been running windows without problems for two months, now i can work without worry.
velis - Thursday, June 23, 2011 - link
I'd hardly call "I have 8 of these and haven't seen the problem so I tend to believe OCZ" putting one's hand into fire. He's just stating his observations.And just so you know:
I started out with OCZ Core 120GB SSD (1st gen crap). Had to replace two of them before support finally figured out that my usage patterns burn out *every* Core SSD in about two months. And I most certainly don't blame OCZ - it was me who was uneducated and bought the wrong disk back then.
So they gave me a Vertex 1 120GB (Indilinx) as a replacement. Been happy with it ever since (sept. 1009).
Additionally I bought 3 Intel X25 80GB G2s. Each and every one of them has given me pain. Not lots of it, but it has. Every once in a while I get a corrupt sector or two, checkdisk runs on it and then I'm fine for the next few weeks. My main computer now needs reinstall because these once-in-a-while events brought the disk (in one year time) into such a state that I had to disable monitor sleep because otherwise the computer won't wake up after it any more. Not to mention other various BSODs and stuff. Luckily I keep my important data on spinning disks and back up to 3 different destinations. Any yes, I know I'm lazy :P
On the other hand, the Vertex never gave me any reason not to like it. In fact - I use it in my work computer and brag about it to all the rest of the company I work for. Luckily for me my bosses don't want to invest into SSDs yet so currently I have the fastest "server" around :D
So, as far as I'm concerned, it's intel who is unreliable, not OCZ...
Jaybus - Friday, June 24, 2011 - link
Why would you say that about the hand in the fire? On almost every one of these reviews, he says something like the following quote from this article.</p><p>"I still believe the Intel SSD 510 is a great balance of performance and reliability. If you want something with an even lower failure rate, there's always the Intel SSD 320 "
Minion4Hire - Thursday, June 23, 2011 - link
I've had a very similar experience to Anand's. Between my own and my immediate family's computer's I am running 2 Intel drives (G2 and 320), 3 OCZ (Agility, Vertex, Agility 2) and Corsair (Force) and have not had the slightest issues with any of them. Of the various computers I have built for other people I have only recently seen a single SSD die, which was an Indilinx Agility from 1.5 years ago. Replaced it with a Vertex 2.People are going to have issues but I, like Anand, don't bother trying to optimize or tweak anything. AHCI enabled, all other drive, BIOS, and OS settings are on their defaults. Everything I've used has worked flawlessly. I'm not saying some people haven't had problems, but is this anger really any different than any other brand rage?
Sandforce is arguably the most popular controller manufacturer atm. Frankly, one of very few options. Obviously some people are going to have problems. And those people are going to be able to find other people who ALSO have those problems. It doesn't mean everyone is having them.....
michal1980 - Thursday, June 23, 2011 - link
Anand is carring water for OCZ.A mystery fat envelop of cash showed up in someones mail box?
So OCZ claims they have a ~1% failure rate, and you just belive them? what manufacture wont lie about that?
P.S. your sample size of 1 or 2, is not good enough to make any claims as to the quality of the product.
The_True - Thursday, June 23, 2011 - link
have you notice that in Anandtech are more Review about OCZ than any other site on the net?lol.... "A mystery fat envelop of cash showed up in someones mail box?
" hahahahah...lol
i mean what is going on Anand??