Sequential Read/Write Speed

Using the 6-22-2008 build of Iometer I ran a 3 minute long 2MB sequential test over the entire span of the drive. The results reported are in average MB/s over the entire test length:

Iometer - 2MB Sequential Read

Both sequential read and write speed are improved over the 5400.6 but we're still not in VelociRaptor territory. The X25-V is cripped in its write performance by comparison but most other SSDs do very well.

Iometer - 2MB Sequential Write

Random Read/Write Speed

This test reads/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.

Iometer - 4KB Random Read

Random read/write performance is abysmal. You can't really make out the numbers here but that's 0.7MB/s for reads and 0.3MB/s for writes compared to 40MB/s+ for the SSDs. It's the poor random access performance that ultimately prevents the Momentus XT from feeling like an SSD most of the time.

Iometer - 4KB Random Write

The Test - Real World First Overall System Performance using PCMark Vantage
POST A COMMENT

120 Comments

View All Comments

  • codedivine - Monday, May 24, 2010 - link

    Another request is to compare performance against a non-hybrid 7200rpm Momentus, which is the same drive minus the flash. That will make the performance benifits clearer. Reply
  • Ralos - Monday, May 24, 2010 - link

    How come a Velociraptor is faster the second time, just after a reboot?

    It should be the same each time, should it not?
    Reply
  • UltraWide - Monday, May 24, 2010 - link

    It's due to the fact that it's a mechanical drive with moving parts. At some times the head might be positioned closer to the data requested resulting in improved performance whereas other times it may have to wait for the next revolution to come to the same data. This creates slight variations in performance. Reply
  • icrf - Monday, May 24, 2010 - link

    I don't think that makes sense. The second and third runs are consistently faster than the first run, and relatively consistent with each other. What causes a purely mechanical hard drive to learn? Reply
  • kmmatney - Monday, May 24, 2010 - link

    I agree - the velocirapter should have parked its read head between boots, so it really shouldn't have been faster between the runs. It must have some tricks. Reply
  • Anand Lal Shimpi - Monday, May 24, 2010 - link

    That would be Windows 7's SuperFetch at work. Reply
  • bitterman0 - Monday, May 24, 2010 - link

    Actually, I was thinking about comparing this hybrid drive to Windows' own methodology of clustering frequently used data on a USB Flash drive (ReadyBoost, is it?). Granted, USB Flash is not as fast as this on-board SLC, and as far as I understand ReadyBoost is using USB Flash for lower latency purposes only. Yet, from what I hear ReadyBoost helps quite a bit in the OS boot process.

    But then I've noticed this reply indicating that SuperFetch was NOT disabled in any of the tests. Well, that pretty much invalidates all results obtained in the tests, don't you think? With 4GB of RAM, SuperFetch overrides much of 4GB of read-only NAND cache, and most likely confuses the cache controller. Have the test configuration had 8GB (or more) RAM, the situation would have been much worse still.

    Is there a possibility to redo the same tests with SuperFetch turned OFF? That would be a synthetic benchmark for Windows Vista and Windows 7 users, but it will be closer to reality for Windows XP, Windows Server and non-Windows users.
    Reply
  • leexgx - Monday, May 24, 2010 - link

    you need to disable the ready boot or ready boost (i got to go out now so not going to check logs to what is doing the boot optimization) as i see 400mb or so of ram been free up after 20-30 secs after desktop is shown Reply
  • jaydee - Monday, May 24, 2010 - link

    Does SuperFetch play the same role on the Momentus XT? Is someone were running WinXP, would we then expect the boot times to be about the same as the Velociraptor, and both of these have the same boot times no matter how many times you rebooted? Reply
  • void2 - Monday, May 24, 2010 - link

    Not exactly. This is the result of ReadyBoot (improved version of "boot prefetch" from Windows XP).

    SuperFetch in Windows 7 is not even active for a few minutes after boot (unlike, alas, Vista).
    Reply

Log in

Don't have an account? Sign up now