Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm using ESXi to host several virtual servers and I'd like to test the speed of various hard drives and protocols. (ie. I'd like to know which drive/setup works best for working operating systems)

I'm going to be comparing:

  • Local SATA drive
  • ZFS - RAID10 (via NFS)
  • ZFS - RAID10 (via iSCSI)
  • ZFS - RAIDZ2 (via NFS/iSCSI)

I'd just like to do some very simple disk benchmarks and I really like the ease-of-use of CrystalDiskMark. Is there a better alternative for simple OS drive benchmarks?

Also, can somebody explain the different numbers of these tests and what they mean?

For example, should I always use a 1 GB test? (in CrystalDiskMark)

Also, what is important for an OS drive and what is important for a storage drive? (ie. 512k speed is most important for an OS drive?)

share|improve this question
Intel released a program called IOMETER a few years ago which would allow you to do this. You are able to install on the ESXi environment and/or run it from a different machine to test a network share. However, it can be taxing on your server while it's running and usually works best with a drive not yet partitioned. – kobaltz Oct 3 '11 at 2:14

Passmark have a Hard Disk(including SSDs) section. You can benchmark your own HDDs arrangement and compare the results with the record list on the site. This is the best that I know because it includes a large range of models and manufacturers.

Another benchmark software that I know is HDTune, however it not have a good record list of user's results.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .