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

  • pSupaNova - Wednesday, October 22, 2014 - link

    No he is correct, Microsoft did not want netbooks to succeed.

    While Intel tried to Gimp their graphics and even stopped Nvidia ION project from improving the graphics situation

    This mistake let tablets have an easier ride and now Nvidia has a remarkable SOC in the K1 thats going to eat Intel alive in the coming years.
  • jabber - Wednesday, October 22, 2014 - link

    Netbooks were awful. They deserved to die. I ended up refusing to fix or work on them for customers. They would bring them to me saying "this is slow!" I would reply "It's because it IS slow!"

    Poor user experience and a waste of money. A novelty that should never have survived longer than a few months.
  • mike8675309 - Tuesday, October 21, 2014 - link

    There were two specific problems. #1 - slow hard drives. #2 - no hardware video acceleration. With the Atom chip and slow drives with no hardware video acceleration, you have what is sitting in my basement and is effectively useless.
  • hojnikb - Tuesday, October 21, 2014 - link

    And its pretty much possible to make one for 150-200$, since windows licenses are free with cheap devices and Intel tablet SoC are selling for peanuts.
    Eeebook x205 is an example of that.
  • inmytaxi@gmail.com - Sunday, November 2, 2014 - link

    The 10 inch laptop form factor is dead for windows, replaced by 10 - 11.6 inch tablets many with removable or flip-able keyboards.
  • sligett - Tuesday, October 21, 2014 - link

    All my users of HP 3105m netbooks (with AMD e-350 processors, and SSD or HDD) chose to move to Chromebooks when they were offered to them. In the schools where I work, a Windows laptop would have to be LESS expensive than a Chromebook to be attractive. In fact, a $200 Chromebook is in many ways more appealing than a free Windows laptop.
  • RU482 - Tuesday, October 21, 2014 - link

    My 12yr old daughter would object to your statement about AMD Brazos. I'll agree, when I bought the Lenovo Netbook that had the E-350 CPU for my traveling notebook/netbook, it was miles ahead of the Atom Z520 based Asus netbook that it replaced.

    Fast forward to 2014, the Lenovo has been handed down to my daughter, who likes to watch Netflix or play flash games on it. She's recently started to complain about how laggy the Lenovo has become. SSD health and free space are still good, wifi signal strength 5 bars....not sure what the excuse is, maybe she's used mom's i5 Haswell laptop too much! I'd say it's time to retire the Lenovo Brazos machine
  • Michael Bay - Wednesday, October 22, 2014 - link

    In any netbook article one can expect a couple of sad AMD fanatics with drivel like that.

    Wake up, Brazos was hot as all hell and didn`t deliver anything Atom couldn`t.
  • waldojim42 - Thursday, October 23, 2014 - link

    Yes, they most certainly did. Video acceleration and useful CPU performance. AMD made netbook CPU's that had significantly higher IPC at a cost of higher power usage.
  • savagemike - Tuesday, October 21, 2014 - link

    It's pretty easy to dual or co-boot linux on a Chromebook, and honestly that would be about the only good reason to upgrade storage to this kind of capacity. It would, of course, provide a whole host of tools/programs which would address some of the drawbacks you mention.
    That makes sense as desktop Linux, like Windows or OSX, are designed for a traditional local storage centric paradigm, where the ChromeOS devices of course are not.
    The one true fault of the ChromeOS file explorer within its own context is the lack of local network integration. You really should be able to see/use shares on a LAN from the files app on a ChromeOS device and you can't.

Log in

Don't have an account? Sign up now