User measurement and tuning of computer hardware and components is largely a lost art - we buy high performance hardware within our budget, and it is what it is. When we require more speed, or more likely just buy to keep up with technological advances, we buy something new.

This came to mind as I was evaluating why the media performance between a couple of home PCs was relatively poor: For whatever reason, streaming media files were stuttering, and photo replication was taking far too long given the quantities of data involved. While I just lived with this for a while (it is just a couple of home PCs), I finally decided to get some real metrics to know what I was dealing with. I decided to look at the numbers from the file system level (rather than measuring just the network itself to catch any high-level issues). Off I went with the fantastic application iozone (which also doubles as an excellent test utility to gauge the performance of various segregated storage systems, which will be another entry coming very soon), analyzing both the wired network and the wireless network.

To make a boring entry even more boring, it turns out that the source computer - running on an nvidia nforce motherboard - was using the nforce chipset networking adapter, with the secondary Marvell adapter disabled. All with stock settings. After running the first set of tests, seeing that the network performance over the 100Mbps network was yielding about 800KB/second of actual transfer (despite the links all reporting a 100Mbps signalling speed), I disabled that network adapter and enabled the Marvell, switching the ethernet cable over.

With that simple action, network throughput suddenly jumped to 12MB/second (the theoretical limit of 100Mbps ethernet). A 15x performance improvement just because I finally decided to measure it and do something about it. Now I think I'll play with the buffer settings to see what further benefits can be gained. Then I'll probably upgrade to 1000Mbps and start again (amazing how inexpensive 1Gbps networking equipment is now).

I always enjoy these exercises because inevitably you go down a path and learn more about a fringe - yet still important - element of our computing world. For instance some interesting details about the file cache of Windows 2000 (which was largely unchanged for Windows 2003).