The Intel SSD 510 Review
by Anand Lal Shimpi on March 2, 2011 1:23 AM EST- Posted in
- IT Computing
- Storage
- SSDs
- Intel
- Intel SSD 510
Random Read/Write Speed
The four corners of SSD performance are as follows: random read, random write, sequential read and sequential write speed. Random accesses are generally small in size, while sequential accesses tend to be larger and thus we have the four Iometer tests we use in all of our reviews.
Our first test writes 4KB in a completely random pattern over an 8GB space of the drive to simulate the sort of random access that you'd see on an OS drive (even this is more stressful than a normal desktop user would see). I perform three concurrent IOs and run the test for 3 minutes. The results reported are in average MB/s over the entire time. We use both standard pseudo randomly generated data for each write as well as fully random data to show you both the maximum and minimum performance offered by SandForce based drives in these tests. The average performance of SF drives will likely be somewhere in between the two values for each drive you see in the graphs. For an understanding of why this matters, read our original SandForce article.
It's a bit unfortunate for Intel that we happen to start our performance analysis with a 4KB random write test in Iometer. The 510's random write performance is only marginally better than the X25-M G2 at 49.7MB/s. The RealSSD C300 is faster, not to mention the SF-1200 based Corsair Force F120 and the SF-2200 based OCZ Vertex 3.
Although not depicted here, max write latency is significantly reduced compared to the X25-M G2. While the G2 would occasionally hit a ~900ms write operation, the 510 keeps the worst case latency to below 400ms. The Vertex 3 by comparison has a max write latency of anywhere from 60ms - 350ms depending on the type of data being written.
Many of you have asked for random write performance at higher queue depths. What I have below is our 4KB random write test performed at a queue depth of 32 instead of 3. While the vast majority of desktop usage models experience queue depths of 0 - 5, higher depths are possible in heavy I/O (and multi-user) workloads:
While the X25-M G2 scaled with queue depth in our random write test, the 510 does not. It looks like 50MB/s is the absolute highest performance we'll see for constrained 4KB random writes. Note that these numbers are for 4KB aligned transfers, performance actually drops down to ~40MB/s if you perform sector aligned transfers (e.g. performance under Windows XP).
Random read performance is just as disappointing. The X25-M G2 took random read performance seriously but the 510 is less than 20% faster than the Indilinx based Corsair Nova. When I said the Intel SSD 510's random performance is decidedly last-generation, I meant it.
Facebook
Twitter
RSS
128 Comments
View All Comments
AnnihilatorX - Wednesday, March 2, 2011 - link
Anand you didn't clarify very clearly what is the difference of naming between 510 series and X-25 G3.The introduction leads me to believe 510 is X-25 G3, or is it not?
Is X-25 G3 going to use 25nm flash instead, so it's another drive? If so, when is the release date of that, and how do we expect its performance compared to 510 in this review? Will the X-25 G3 uses a custom controller?
strikeback03 - Thursday, March 3, 2011 - link
As the article stated, the G3 (whatever it is officially called) will be a lower-performance part whose aim is to bring lower prices and better reliability to more mainstream segmentsMrStromberg - Wednesday, March 2, 2011 - link
So I've been waiting for quite some time now for the new intel drives since the reviews in the past about how reliable and how long life an SSD has, promising that all of this would be better with these new generation drives. Although now I am faced with the potential "problem" of a 3Gbit bottle neck in mi macbook pro and as mentioned in the review above "these next-generation SSDs not only use 6Gbps SATA, they really need it." So where does that leave me? Should a go for a cheaper older drive which might be less reliable in the long run (but nobody really knows right?) or buy a new generation drive which might be suffering from a bottleneck? I didn't really understand why the new drives really need the 6Gbit SATA to function well? Can someone please explain or give me some advice.thank you
Denithor - Wednesday, March 2, 2011 - link
Best right now would probably be an Intel G2 drive. Fast enough and very good durability.The next generation drives aren't really going to perform much better on a 3Gb SATA port than the current generation already does, plus you have the worse durability inherent in the 25nm NAND chips.
Nentor - Wednesday, March 2, 2011 - link
Wow, so Intel has a gap in it's road map and goes the 3rd party route and brings us a product that is essentially slower than a consumer product (Vertex 3) from another manufacturer.They could have gone Sandforce, why not?
mateus1984 - Wednesday, March 2, 2011 - link
http://forums.hexus.net/general-discussion/199892-...Drag0nFire - Wednesday, March 2, 2011 - link
Would love to see you run a traditional hdd through the new bench so we can see what sort of real world improvements can be expected from making the switch to an SSD...Great article. I'll be waiting for the x25-m G3. Keep on Intel about this!
wheel - Wednesday, March 2, 2011 - link
Thanks for the review Anand, but I am a little disappointed that there are a lot of synthetic benchmarks but no real world tests.By "real world tests" I mean separate tests for: booting into Windows; loading a web browser with 25 tabs of saved web page documents on the HDD; starting Star Craft and loading a map; copying a large amount of files to itself; running a batch Photoshop image transform job; starting IE6 from a stopped Windows XP Mode VM and opening a complex web page hosted on the local disk; running a intensive anti-virus scan on a specific (large) folder etc.
I know PC Mark and SysMark are meant to represent these real world test, but as individual consumers we have different usage profiles and by breaking down the results into individual tests we can better work out which drive is most appropriate for us, instead of studying the synthetic tests and making an educated guess.
Cheers,
Ian
Boogaloo - Wednesday, March 2, 2011 - link
Seconded.A lot of people are complaining about the performance of the drive, and I'd like to know how much of a difference it actually makes in real world scenarios. If this drive comes within 10 ms of a vertex 3 loading up starcraft 2, then who cares?
iwod - Wednesday, March 2, 2011 - link
If you read carefully the Anand Benchmarks does exactly just that.