The Intel SSD 320 Review: 25nm G3 is Finally Here
by Anand Lal Shimpi on March 28, 2011 11:08 AM EST- Posted in
- IT Computing
- Storage
- SSDs
- Intel
- Intel SSD 320
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.
Random write speed is improved compared to the 510 thanks to Intel's controller, but we're only looking at a marginal improvement compared to the original X25-M G2.
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:
Random read performance has always been a strong point of Intel's controller and the 320 is no different. While we're not quite up to C300 levels, the 320 is definitely competitive here.
Sequential Read/Write Speed
To measure sequential performance I ran a 1 minute long 128KB sequential test over the entire span of the drive at a queue depth of 1. The results reported are in average MB/s over the entire test length.
Without a 6Gbps interface the 320's performance is severely limited. Compared to other 3Gbps drives the 320 is quite good here though.
Read performance is at the top of the chart for 3Gbps drives. I wonder how far Intel would've been able to push things if the 320 had a 6Gbps controller.
Facebook
Twitter
RSS
194 Comments
View All Comments
Chloiber - Tuesday, March 29, 2011 - link
True.It is true, thet the REAL capacity of flash drives is 2 based. The NAND chips are.
So a 120GB drive has in reality 128GB of flash.
So its 120/128 % spare area. The 300GB version has also 300/320 % spare area (which is exactly the same).
Anand is confusing things. The user gets 300GB, as he gets 300GB when buying a HDD. Windows on the other han is showing us "GiB" not "GB". But it's not a real difference in size. 74.5GiB EQUALS 80GB. It's the same thing. Compare the BYTE numbers if you wanna be sure, not the KB/MB/GB/TB numbers.
I'm actually shocked that this still gets confused.
overzealot - Tuesday, March 29, 2011 - link
RAM was not the only thing that was calculated using binary pseudo-metric prefixes. Perhaps you aren't old enough to remember the days before kibibytes, when all computer disks and tapes were measured as such.noblemo - Wednesday, April 6, 2011 - link
Conversion from GB to GiB:320 / 1.024^3 = 298 GiB
Subtract 6.25% spare area:
298 x (1-0.0625) = 279 GiB
MeanBruce - Monday, March 28, 2011 - link
You can blather all the technostats you want 25nm, who cares, didn't change a thing! My next ssd, would never have said this a year ago, looks like Corsair Force GT! Read/Write 500/500 is all you need to say!;)zanon - Monday, March 28, 2011 - link
Granted, this isn't a stunning offering. But one thing I do look forward to is that I think we will finally start to see updated filesystems start to appear in the near future. For example, ZFS appears as if it will at last appear as a full Mac OS X file system via Z-410 this summer.One of the features of modern filesystems is full filesystem level compression and encryption (which really is where such features belong). I will be looking forward to (hopefully) seeing you test how this affects the SSD scene. My principle concern with Sandforce's strategy in the back of my head has always been this: that sooner or later, OS makers or someone will finally get with it and make full compression standard in the FS. At that point, the "worst case" scenario of fully random data will become the *only* scenario. That still leaves a (huge huge) legacy market, and likely time to adapt, but I do wonder if it will shake up the SSD scene at all once again.
overzealot - Tuesday, March 29, 2011 - link
I don't agree. If the controllers are powerful enough to do encryption and compression in real-time, then it should still be done at the disk level.You can still encrypt/compress in your OS as you please, but I like having performance.
PS, not dogging on ZFS, I use it all the time with openindiana.
marc1000 - Monday, March 28, 2011 - link
vertex 3 is not already on market???http://www.amazon.com/OCZ-Technology-Vertex-2-5-In...
aork - Monday, March 28, 2011 - link
That's for pre-order. Notice "Usually ships within 1 to 2 months."piquadrat - Monday, March 28, 2011 - link
Is it sufficient, security wise, using only max. 8 characters ATA password against thieves?One program, I sometimes use, MHDD offers ATA password reset option.
If someone can bypass ATA pass so easily what all this AES128 is for?
Could someone explain this matter to me?
DesktopMan - Monday, March 28, 2011 - link
The password is used to generate the encryption key, much like how software products such as TrueCrypt does it.The max length of ata paswords is 32 which should be more bits than the actual key, depending on character set. 8 is not much though, depending on how these drives deal with brute force attacks.
Old drives with ata passwords are just enabled with the password, which can be circumvented with master passwords or firmware commands in some cases.