Performance Consistency

Performance consistency tells us a lot about the architecture of these SSDs and how they handle internal fragmentation. The reason we do not have consistent IO latency with SSDs is because inevitably all controllers have to do some amount of defragmentation or garbage collection in order to continue operating at high speeds. When and how an SSD decides to run its defrag or cleanup routines directly impacts the user experience as inconsistent performance results in application slowdowns.

To test IO consistency, we fill a secure erased SSD with sequential data to ensure that all user accessible LBAs (Logical Block Addresses) have data associated with them. Next we kick off a 4KB random write workload across all LBAs at a queue depth of 32 using incompressible data. The test is run for just over half an hour and we record instantaneous IOPS every second.

We are also testing drives with added over-provisioning by limiting the LBA range. This gives us a look into the drive’s behavior with varying levels of empty space, which is frankly a more realistic approach for client workloads.

Each of the three graphs has its own purpose. The first one is of the whole duration of the test in log scale. The second and third one zoom into the beginning of steady-state operation (t=1400s) but on different scales: the second one uses log scale for easy comparison whereas the third one uses linear scale for better visualization of differences between drives. Click the dropdown selections below each graph to switch the source data.

For more detailed description of the test and why performance consistency matters, read our original Intel SSD DC S3700 article.

Micron M600 256GB
Default
25% Over-Provisioning

Performance consistency is not very good, but that was expected as we are dealing with a DRAM-less design. With insufficient cache for the NAND mapping table, it is very hard to manage the table efficiently in the steady-state environment, which results in decreased performance. The good news is that Chrome OS is extremely light and a scenario like this is basically impossible, so the data here is merely an interesting piece of information.

Micron M600 256GB
Default
25% Over-Provisioning

 

Micron M600 256GB
Default
25% Over-Provisioning

TRIM Validation

To test TRIM, I filled the drive with sequential 128KB data and proceeded with a 30-minute random 4KB write (QD32) workload to put the drive into steady-state. After that I TRIM'ed the drive by issuing a quick format in Windows and ran HD Tach to produce the graph below.

It seems that TRIM is working, although there is an odd drop in read performance after the ~140GB mark.

Managing Storage in Chrome OS AnandTech Storage Bench 2013
Comments Locked

67 Comments

View All Comments

  • abianand - Thursday, October 23, 2014 - link

    Thank you...very stupid of me posing a question without reading the article fully. First time I've done that.

    Spending around 75-80 for an ssd upgrade negatively affects the value proposition of the chromebook, which in the first place is questionable (to my purpose)
  • AnnonymousCoward - Thursday, October 23, 2014 - link

    5 pages talk about performance, yet there's no inkling on actual user performance. You say "the performance difference is tremendous" - back that up with boot time or app launch time! Just looking at synthetics is idiotic.

    Realtime power consumption is less meaningful than total power consumed for some task. Faster transfer rates use more power while it's transferring but operate for less time.
  • cutterjohn - Saturday, November 1, 2014 - link

    Also remember AMD is AWFUL wrt power usage.

    I recently bought a C720/i3/4GB version and replaced the shipped SSD w/128GB my digital -> ~100GB available. I suspect that part of the SSD space is walled of for OS reasons as there is a factory restore(in Android parlance) option which means a clean copy is stored somewhere, probably on it's own partition, and part of the space is probably also saved as misc space(I forget what they call it for SSDs, spare?).

    Anyways ATM I've also croutonized it w/ xfce/Ubuntu 14.04, there's also croagh which supports more distros, e.g. ARCH, and chrubuntu which just replaces chrome os IIRC(wasn't ready to do that and probably won't*.)

    * I used chrome os exclusively for a few days(waiting for SSD delivery) and it was quite fast and power efficient, regularly giving me ~8h of runtime. It was able to play netflix & amazon instant video OOB, but I haven't tried again since dev moding it and croutonizing it. (I've seen comments that dev mode removes netflix streaming capability, but they were old.)

    Even with chroot Ubuntu/xfce it's pretty quick. A fair number of steam games are available for it, and I've just recently installed wine, but have not gotten around to trying out and windows apps/games yet although there are some videos of even skyrim(low settings although the videographer claims medium does OK as well) running on a celery chromebook.

    Personally, I'd like an even smaller chromebook, more in the original netbook form factor, in which case 1280x720 or the existing 1366x768 would be quite dense although cramping the keyboard. I mainly picked it up for size and batt runtime as even my Sager 7330 is a bit bulky at times(13") and batt runtime is usually only ~3h. I also wouldn't have minded an i5 option as well and slotted RAM...
  • omegajimes - Tuesday, November 4, 2014 - link

    Do people actually use Chromebooks with ChromeOS? I thought the only reason to pick these up were to load lightweight linux distros like Mint or Elementary on them.
  • leoku - Wednesday, November 5, 2014 - link

    I think the Commodore C64 all-in-one in the great grandpa's era is more fun than these netbooks and chromebooks, plus you can program it right out of the box. A Commodore C64 is more expensive than Chromebook nowadays, though.
  • leoku - Wednesday, November 5, 2014 - link

    Given the fact, I think the SSD upgrade for a Chromebook is a waste of money.
  • dareo - Sunday, November 30, 2014 - link

    I’ve just ordered a Lenovo S540 and it comes with a 1 TB HDD (5400 rpm) + a 16 GB M.2 SSD for caching. This leads to several questions related to the MyDigitalSSD 128GB Super Boot Drive:

    1. Could I swap out the 16 GB M.2 that comes with the Lenovo, replace it with the MyDigitalSSD 128GB Super Boot Drive, install the OS and apps on it and use it as the primary boot drive?

    2. What’s the difference between the MyDigitalSSD 128GB Super Boot Drive and the MyDigitalSSD SC2 Super Cache 2 42mm SATA III 6G M.2 NGFF M2 SSD Solid State Drive?

    3. How reliable is this M.2 SSD in comparison to a Crucial MX100 or a Samsung EVO?

    4. Would it just make more sense to pluck the M.2 and HHD from the computer and install a 500 GB Crucial MX100 or a Samsung EVO?

    Thanks in advance for any feedback / advice on this.

Log in

Don't have an account? Sign up now