Performance vs. Transfer Size

ATTO is a useful tool for quickly measuring the impact of transfer size on performance. You can get the complete data set in Bench.

These charts give us a great look at the various graduations of performance as we scale up NAND die count within the M500 family. The 480/960GB drives perform identically, while the 120/240GB drives show significant steps down in max sequential read performance.

Write speed is a bit closer between all of the M500 capacities, but none approach the peak performance of Samsung's 840 Pro.

 

Random & Sequential Performance AnandTech Storage Bench 2011
Comments Locked

111 Comments

View All Comments

  • blackmagnum - Tuesday, April 9, 2013 - link

    How times have changed. SSDs now have better bang-for-the-buck than hard disk drives. No noise, low power, shock resistance... the works.
  • Flunk - Tuesday, April 9, 2013 - link

    That's not quantitatively true. 2TB hard drives are available for about $100 which is 0.05/GB no SSD can match that.

    SSDs have better power usage, performance, shock resistance but they lag in capacity.
  • ABR - Tuesday, April 9, 2013 - link

    Actually they tend NOT to have better power usage, at least when compared against 2.5" laptop hard drives. But everyone thinks they do anyway since it just seems like a purely electronic device should use less energy than a mechanical one.
  • akedia - Tuesday, April 9, 2013 - link

    I don't believe that's correct. Sure, sometimes, some SSDs in some usage scenarios might use more, but it's generally correct that SSDs use less power than even 2.5" HDDs. Below I've linked to recent (within the last year) StorageReview.com reviews for two current generation examples, the WD Scorpio Blue and the Samsung 840 Pro. The SSD beats the SSD on all measures other than writing, which over the course of time is unlikely to tip the scales, and even then they have to note that the system was tested in a desktop which didn't have DIPM enabled. So this is a worst-case for the SSD, without all of its power-saving features enabled, and it still comes out generally on top, while providing vastly superior performance on all measures.

    It's not true that ALL SSDs beat ALL HDDs at ALL times for ALL usages in ALL circumstances, but it's also not true that 2.5" HDDs have better power usage in general. They don't. And that's without even considering how much less time such a higher performing device would take to read or write a given amount of data, spending much less time out of power-sipping idle. Cheers.

    http://www.storagereview.com/western_digital_scorp...

    http://www.storagereview.com/samsung_ssd_840_pro_r...
  • akedia - Tuesday, April 9, 2013 - link

    *edit

    The SSD beats the HDD on all measures other than writing, it doesn't beat itself. *facepalm*
  • ABR - Tuesday, April 9, 2013 - link

    The SSD link you site, together with another review on Tom's Hardware, report
    very different values for power usage than most places I've seen. For example,
    here on Anandtech:

    http://www.anandtech.com/show/6328/samsung-ssd-840...

    Generally averaging 3-5 watts, whereas good HDDs are in the 1.5-2.5 range. It would be good to know the reason for the discrepancies. It does seem that smaller processes are starting to help the SSDs catch up though.
  • tfranzese - Tuesday, April 9, 2013 - link

    You do realize that SSD's can get their work done quicker and get back to idle much faster than any mechanical drive? Unless you're looking at a SSD that has horrible idle power characteristics there's little hope in hell for a HDD to compete as far as power efficiency goes.
  • ABR - Tuesday, April 9, 2013 - link

    What I realize is that there is a lot of hand-waving and warm fuzzy thinking in this area, but few hard numbers. The ones that I *have* seen tend to suggest SSDs are still catching up in power efficiency.
  • MrSpadge - Tuesday, April 9, 2013 - link

    +1

    Consuming about as much power (give or take a few 10%) for one or two orders of magnitude less task completion time results in one to two orders of magnitude less energy consumed to complete the task. And that's what really counts, for the wallet and the battery.
  • mayankleoboy1 - Wednesday, April 10, 2013 - link

    Instead of "Power usage" , lets see the "Energy usage" of the whole system, (that is powerused*time)
    I strongly suspect that SSD's will easily beat any HDD here.

Log in

Don't have an account? Sign up now