Comments on: Comparing the I/O Performance of 2 or more Virtual Machines SSD, SATA & IOmeter

By: ecards_guy

ecards_guy — Fri, 15 Jan 2010 23:15:16 +0000

It’s very hard to believe that a good sata drive will outperform a good SSD at anything, and I don’t think you have proven that here – we must know the details of the drive you were using.

Are you meaning to say that a bad or old model SSD is not as good as SATA for VMs? That’s easy to believe.

It would be great to see results for Intel’s G2 SSD, or any Barefoot controller based drive.

SSD drives older than the ones I mention don’t represent the current state of the SSD market, which changes very quickly.

By: It’s voting time.. « Virtualization, Windows, Infrastructure and all that “stuff” in-between

Mon, 04 Jan 2010 22:04:30 +0000

[...] Comparing disk I/O of virtual machines on SSD and SATA disks with IOmeter [...]

By: Gabrie

Gabrie — Fri, 18 Dec 2009 12:07:41 +0000

Great post, going to play with new IX4-200D today

By: So you bought an equallogic san, now what…. part three « Michael Ellerbeck

Tue, 08 Dec 2009 19:57:55 +0000

[...] http://vinf.net/2009/11/19/comparing-the-io-performance-of-2-or-more-virtual-machines-ssd-sata-iomet… [...]

By: Duncan

Duncan — Thu, 19 Nov 2009 10:10:52 +0000

that’s one of the reasons for not selecting SSD in my new desktop….

By: vinf

vinf — Thu, 19 Nov 2009 07:15:01 +0000

Hey Chad, thanks for that; doing some research into this so that’s useful

By: Chad Sakac

Chad Sakac — Thu, 19 Nov 2009 03:17:49 +0000

Where SSD will shine is small (4K-8K) random IO – ergo look for workloads that drive tiny bandwidth (MBps) and massive throughput (IOps).

Where SATA will shine is large (256K+) sequential IO – ergo look for workloads that drive massive bandwidth (MBps) and tiny throughput (IOps).

Chad