Sign in with
Sign up | Sign in

Results: Random Performance

SanDisk Extreme II SSD Review: Striking At The Heavy-Hitters
By

Iometer is still our synthetic metric of choice for testing 4 KB random performance. Technically, "random" translates to a consecutive access that occurs more than one sector away. On a mechanical hard disk, this can lead to significant latencies that hammer performance. Spinning media simply handles sequential accesses much better than random ones, since the heads don't have to be physically repositioned. With SSDs, the random/sequential access distinction is much less relevant. Data can be put wherever the controller wants it, so the idea that the operating system sees one piece of information next to another is mostly just an illusion.

4 KB Random Read

Plextor's M5 Pro and the SanDisk drives offer similar performance. Throughout the capacity range, the Extreme IIs are competitive. The 120 GB model isn't as strong, but it's almost exactly as fast as the 240 GB Seagate 600.

The 240 GB and 480 GB Extreme IIs don't quite hit 100,000 IOPS, but there's no shame in 94,000 and 91,000 IOPS, either.

4 KB Random Write

And then things seem to go pear-shaped. A glance at the above chart makes it clear that SanDisk's drives aren't living up to their specifications. Shouldn't they be hitting 80,000 IOPS or so?

The explanation is relatively simple. We test with random data over a 16 GB LBA space. Industry-wide, most consumer-oriented SSD tests are limited to 8 GB. Now, this doesn't matter most of the time. Hard drives are especially sensitive to LBA active ranges, since spinning platters and floating heads need more time to move when the data you request is physically farther away. Solid-state storage obviously isn't subject to the same limitation, though some SSDs are more sensitive to changes in LBA ranges than others. The difference just usually isn't so profound.

Using the 240 GB Extreme II, we can demonstrate this idiosyncrasy. Starting with 1 GB of sectors and graduating to the entire LBA range, the drop in performance is substantial by the time we get to 16 GB. There are technical reasons why this might happen to a lesser degree with other SSDs, but it looks like the implementation of nCache can result in slower random writes at high queue depths over a large number of LBAs. It's possible that a tradeoff exists between writes that can be cached and writes that exceed the cache's capacity, hurting performance when the cache is full and improving speed when nCache can effectively handle smaller random writes.

Is this a problem? In a word, no.

Typically, random workloads bombarding the entire drive are considered enterprise-oriented. Consumer usage just doesn't match that profile. Random writes are more typically limited to smaller areas, and the amount of writing is exceptionally light. The fact of the matter is that SanDisk's Extreme II was designed for desktop workloads. Wringing the last few drops of performance from an interface-limited SSD means taking steps to improve one area at the expense of others.

The trade-off seems fair. The Extreme II is less useful for a selection of some enterprise applications, but is better as a boot and desktop application drive. We can live with that.

Besides, the Extreme IIs aren't even as bad off as they might appear. Consider the above 4 KB write saturation test at a queue depth of 32. Sure, the drives start well off of their highs. But after the SSDs are filled and garbage collection is in full swing, SanDisk's solutions aren't any worse than competing models. In some cases, they're even better. Performance levels off and ranges from 7,000 to 10,000 IOPS, depending on size. If nothing else, that's competitive. Regardless, there just aren't any reasons why you'd ever write like this on a gaming rig.

Ask a Category Expert

Create a new thread in the Reviews comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 40 comments.
This thread is closed for comments
  • 0 Hide
    Someone Somewhere , June 6, 2013 10:33 PM
    Where's the 840/840 Pro?
    Also, you appear to have put one of the labels back on the wrong way round.
  • 0 Hide
    awez , June 6, 2013 10:50 PM
    My thoughts exactly, where's the 840 and 840 pro?
  • 0 Hide
    boulbox , June 7, 2013 12:34 AM
    I have always been a fan of Sandisk SSDs, can't wait until to try this out in someone else's build as they usually sell their products that is very acceptable for budgets.
  • 0 Hide
    Dixevil , June 7, 2013 12:44 AM
    heavy hitters with no 840pro
  • 0 Hide
    slomo4sho , June 7, 2013 12:52 AM
    I am also curious about the selection of the comparative models. Having the Extreme (not II) in the charts for comparison between the two generations would have been a welcomed addition along with the inclusion of the 840 series.
  • 1 Hide
    flong777 , June 7, 2013 12:59 AM
    I know a lot of people have already pointed this out but can't Tom's Hardware afford a damn 256 GB 840 Pro? I mean come on, it is the fastest SSD on the planet right now.
  • 0 Hide
    raidtarded , June 7, 2013 1:02 AM
    Seriously, what is the point of this article? The fastest car in the world is as Yugo if you dont test against a Lamborghini.
  • -1 Hide
    teh_gerbil , June 7, 2013 1:04 AM
    Why are there 2 of your most recent SSD reviews lack the Samsung 840/Pro? Are you being paid by the respective companies to avoid using them, as for both SSD's, as per other reviews I have read, the 840 Pro cr@ps all over both of them, but due to your lack of them, they're both top of your benchmarks! Very very bad benchmarking.
    http://www.tomshardware.com/reviews/vertex-450-256gb-review,3517.html
  • -1 Hide
    merikafyeah , June 7, 2013 3:47 AM
    Want an 840 Pro comparison and far more in-depth review?
    See here: http://www.anandtech.com/show/7006/sandisk-extreme-ii-review-480gb

    It's Anand's new favorite SSD, and based on the results, I'm inclined to agree.
    It's peak performance is right up there with the 840 Pro, but what's really extreme is the drive's consistency. It's performance when the drive is close to full is unmatched.

    There are no high peaks accompanied by low valleys in performance when it comes to the Extreme II. It's pretty much smooth and fast sailing all the time, which in my book, places the Extreme II a step above the 840 Pro. The 840 Pro would have to be at least $30 cheaper than the Extreme II for me to even consider it over the Extreme II.
  • 0 Hide
    JPNpower , June 7, 2013 6:25 AM
    Why is the 840 Pro the fastest SSD on the planet? It has its share of drawbacks, and is slower than the OCZ Vector, and the Plextor M5 Pro Xtreme on many benchmarks. Don't make broad statemets that aren't always true.
  • 0 Hide
    JPNpower , June 7, 2013 6:33 AM
    To Toms,
    The "Heavy hitters" for modern SSDs include the fastest SSDs on the market right now, which are The Plextor M5 pro Xtreme, the OCZ Vector and Samsung 840 pro. Of these, you have only included the OCZ, and the slower version of the Plextor. Also, you have also included the old Crucial m4, which is a good drive, but old, and not one of the heavyweights now. At least include the improved "M500" version. I also find it confusing why you include the older Samsung 830.
    These are minor points though. Thank you for the great comparison. I look forward to more storage comparisons
  • 0 Hide
    Branden , June 7, 2013 7:02 AM
    you call the article "striking at the heavy-hitters" yet you don't compare it to THE heavy-hitter: the samsung 840 pro.
    that single omission itself made this review critically flawed.
  • 0 Hide
    povu , June 7, 2013 7:11 AM
    I'm still using a Sandisk Sansa Fuze mp3 player, good stuff.
  • 0 Hide
    bucknutty , June 7, 2013 7:18 AM
    About a year ago I got a Sandisk extreme 120gb on sale for $90. I knew it was not the fastest drive or the most high-end drive, but the price was right. It has been running 12-15 hours a day every day for the past year and it works great. Its fast for video editing and loading video games, and that's all I wanted it. Sandisk has put out 3 updated firmwares in that time as well as a little, ssd health program, so you can monitor your read writes, update the firmware or check the ssd for errors. I feel that I got a great value for my $90.
  • 0 Hide
    foolishone , June 7, 2013 8:55 AM
    Tom's you really need to stop referring to these components as SandForce. It was acquired by LSI 17 months ago.
  • 0 Hide
    cangelini , June 7, 2013 12:29 PM
    We're working on getting Christopher multiple capacities of the 840 Pro to add to his library of drives. We have nothing against the 840 Pro. In fact, the rest of our staff is using them as our reference for 2013. The fact that Christopher doesn't have one is simply an artifact of him recently coming on-board as our consumer SSD editor.
    Thanks,
    Chris
  • 1 Hide
    computertech82 , June 7, 2013 3:40 PM
    I would REALLY like to see RELIABILITY, NOT about just speed reviews. Like the 840 pro that has MANY post on newegg about dying drives (and other models/makes as well). CORSAIR is the only one that doesn't but they also don't have that many post.
  • 0 Hide
    danwat1234 , June 8, 2013 1:04 AM
    Why wasn't the Intel X25-m G2 SSD in these benchmarks? It's still a very good reliable drive and I'm interested in how it compares in random reads/writes.

    What we are seeing is stagnation. We have a great Marvell controller, Indilinx Barefoot 2 controllwer and a solide Sandforce 2000 series controllered SSDs.
    I'm waiting for the next generation, maybe for the Sandforce/LSI 3000 series controllers that can do 200,000 IOPS! Google it. Though that drive was using a PCIe 4x interface rather than SATA but it was in the 2.5" drive form factor.
  • 0 Hide
    Someone Somewhere , June 8, 2013 3:13 AM
    200K IOPS is 800MB/s of 4K transfers. Not going to happen on SATA 6Gb/s which is 600MB/s MAX, including overhead, after 8b/10b encoding.
  • 0 Hide
    danwat1234 , June 8, 2013 5:46 AM
    I agree with you but it would finally be an SSD that would saturate the SATA 3 interface with Tomshardware's 'Storage bench 1.0'. Right now the fastest SSD maintains an average speed of 268MB/s. Probably significantly faster real world desktop traces too and PCmark.
    Maybe increase queue depth of 1 4KB random reads and write speeds too. So far I've only seen as high as about 30MB/s 4K random read with a queue depth of 1 on Crystaldisk Mark.
Display more comments