[ZzBloopzZ]

The FAQ seems to be rather outdated. Which of the two are most accurate for measuring HD benchmarks overall? I know they don't measure real-world performance but gives somewhat an idea. Also, are they reliable to measure RAID 0 performance now?

I have been out of the scene for a while, but it looks like you guys are using HD Tune the most these days.

Thanks!

[Fedor]

It's funny that you should say they don't measure real-world performance but also ask which is most accurate

For the record, they measure sequential read/write speeds and access times. This does not translate into real-world performance, at all. It gives a good foundation for the drive but its the firmware that then actually gives performance. Take the Seagate Cheetah 15K.5 - it trounces any 7200rpm SATA disk or Raptor in both sequential rates and access times, and yet in reality is considerably worse for desktop use. Obviously I used a SCSI example to really illustrate the issue, but the same goes for 7200rpm SATA drives - a lot of them can have similar transfer rates and access times but ridiculously different real-world performance.

So basically, those two utils aren't worth much at all. They are both accurate though. HDTune Pro allows some more elabourate testing now, as in you can do drive raw speed testing as well as filesystem testing.

For measuring RAID performance in terms of sequential reads and writes, ATTO is still the way to go.

[noegruts]

Fedor, on Mar 11 2008, 03:12 AM, said:

For measuring RAID performance in terms of sequential reads and writes, ATTO is still the way to go.

You'd suggest ATTO instead of IOmeter?

[Fedor]

noegruts, on Mar 11 2008, 08:19 AM, said:

Fedor, on Mar 11 2008, 03:12 AM, said:

For measuring RAID performance in terms of sequential reads and writes, ATTO is still the way to go.

You'd suggest ATTO instead of IOmeter?

For a simple sequential test? Sure why not.

[ericg]

HD tune crashes my system. I use hd_speed to check read speed, very simple and tiny.