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.

Iometer - 4KB Random Write, 8GB LBA Space, QD=3

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:

Iometer - 4KB Random Write, 8GB LBA Space, QD=32

Iometer - 4KB Random Read, QD=3

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.

Iometer - 128KB Sequential Write

Without a 6Gbps interface the 320's performance is severely limited. Compared to other 3Gbps drives the 320 is quite good here though.

Iometer - 128KB Sequential Read

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.

Spare Area, 3Gbps Only, AES-128 AnandTech Storage Bench 2011: Much Heavier
POST A COMMENT

193 Comments

View All Comments

  • LeTiger - Monday, March 28, 2011 - link

    If Intel's not careful... they will price and performance themselves out of the very market they sought to create in the first place... Reply
  • nexox - Monday, March 28, 2011 - link

    Meh, this is the cheapest SSD with some form of capacitor backup for volatile data - to get something similar you need to go to a SandForce 1500 or 2500 controller, and probably expect to pay 2-3x as much. Most people aren't too concerned with this, but it's an essential feature, no matter what the manufacturers say.

    It's also the only SSD I've seen that doesn't exhibit random latency spikes, that combined with the power fail protection means that it's the only SSD I'd consider buying.
    Reply
  • cactusdog - Tuesday, March 29, 2011 - link

    At least with Intel you are guaranteed to get best quality nand and reliable performance over the long term. Intel have also done the right thing by rebranding 25nm drives, even though in Intel's case performance is a little better than their 34nm drives.

    With OCZ, theres no way of knowing which nand you will get and you could end up with second rate nand not intended for high performance SSD's.

    Intel's performance is very reliable over the long term, whereas OCZ/sandforce 3GB/s drives are known to slow to a "settled state" especially with smaller drives.

    Intel will sell plenty of these drives because they are reliable and trustworthy.
    Reply
  • Griswold - Tuesday, March 29, 2011 - link

    Exactly. Not everybody plays benchmarks all day long and doesnt care one bit about reliability of the storage system. Reply
  • wumpus - Tuesday, March 29, 2011 - link

    And just how many filesystems automatically write everything the instant write() is called? It is a bad idea for rotating media, and will take awhile before filesystems optimized for SSDs show up (Microsoft has been promising a database centric filesystem for the "next windows" since NT was new. Maybe someday.) Also, NTFS has corruption issues with standard hard drives, this isn't going to help that reliability much anyway. You need backups, a real UPS, and RAID (in roughly that order) if you care about data reliability.

    If intel was willing to compete largely on reliability, they could double the price (if data on SRAM competition is even more expensive), as such a feature is worth much more than any amount of hardware cost to a large number of customers. Don't expect to get it without major software surgery (and yes, some of those customers have the software).
    Reply
  • seapeople - Tuesday, March 29, 2011 - link

    I don't understand all the negative feedback on this drive. For the 90% of us who have 3Gbps SATA, compared to the Vertex 3 this drive is:

    ~25% slower than Vertex 3
    ~20% cheaper than Vertex 3

    Why is that so terrible, the Vertex 3 is an amazing performance feat, and under common mainstream conditions this drive is about equal on a price/performance basis.

    Meanwhile, this drive gives us 20% more performance than the g2 generation while being ~20% cheaper. This is far from sad.
    Reply
  • dagamer34 - Tuesday, April 05, 2011 - link

    Except the MSRP for a 120GB Vertex 3 is cheaper than what an equally sized Intel SSD 320 series drive costs. That's simply laughable. Higher price for worse performance? Throwing up the "reliability" card is a red herring when you consider that the 25nm process is brand new, you don't get to state that for any technology until it's been around on the market for quite some time (otherwise, it's just name recognition only, and that worked so well for Sandy Bridge chipsets). Reply
  • seapeople - Thursday, April 07, 2011 - link

    You're on crack. A 120 GB Vertex 3 is listed as 249.99 versus 209.00 for the 120 GB Intel SSD 320.

    Furthermore, SSD reliability issues generally arise from controller/firmware nuances and bugs, not the new process node. It sounds like the SSD 320 is basically the g2 with new flash and a few features unlocked, so it's reasonable to expect similar controller reliability. Apart from that, basically everyone uses the same flash anyway, so if the Intel drive dies because of the new flash process then so will the Vertex.
    Reply
  • andreyu - Thursday, July 21, 2011 - link

    1. you're blind or on crack
    2. you bought a ocz/corsair ssd cause you wanted speed but returned it for problems and now you're mad on intel?
    3.so sorry 4 u

    120 GB Intel SSD 320 is CHEAPER than 120 GB Vertex 3
    Reply
  • TonyB - Monday, March 28, 2011 - link

    lol Reply

Log in

Don't have an account? Sign up now