I finally got around to using a Kill-a-Watt meter to measure energy usage for my Penguin Computing Linux workstation and my new Mac Mini. Jorden Mauro, a co-op who’s been working in my group for a few months now, helped out. We wound up using his MacBook Pro for additional comparisons for reasons which will be evident.

At idle, the workstation uses 89W of power, while the Mac Mini uses 18W and the MacBook Pro uses 22W. That’s a significant savings.

When I built Zoran’s IPS product from scratch on the Linux workstation, the energy usage shot up to a 100-125W range. Since I still don’t have IT’s blessing to have the Mac Mini inside the company’s firewall, I could not use it to do a direct rebuild. Instead, Jorden used his MacBook Pro to do a complete Linux kernel rebuild. At full load, the energy usage peaked at 40W, over half of what the workstation pulls at idle.

To test the Mini at load, we put in a DVD and measured the energy usage. While playing a DVD, the Mac Mini used roughly 36W of power.

That is a pretty significant savings. Now, it should be said that my Linux workstation is a 5 year old machine. It is running a Pentium 4 processor and lacks the power savings features of today’s Core 2 Duo chips. Still, the energy savings coupled with a faster, more capable processor is nice bonus.

This entry was posted on Wednesday, November 19th, 2008 at 3:05 pm and is filed under Hackery. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.