Are you ready for servers that you can put in the compost heap? The Open Compute Project has challenged students at Purdue University to develop a biodegradable server chassis, making data center technology more sustainable.
“We think there’s enormous potential here, as servers are sometimes replaced as often as every 2-3 years,” writes John Kenevy on the Open Compute blog. “And even though the steel in those server chassis is usually recycled, we think it’s worth exploring designs that retain the needed resiliency but push the boundaries of sustainability.”
Even recycling steel generates waste. So what would happen if these chassis could be placed in compost instead? To find out, students from Purdue’s College of Technology entrepreneurship program, Tech Ventures, will tackle the Open Compute challenge in the spring 2013 semester. Students will review relevant research and then break into teams to develop proposals. The winning teams will receive funding to build a prototype.
Purdue’s Interest in Sustainable Containers
Many containers for food and snacks are now available in biodegradable packaging. A server chassis will need to be sturdy enough to support components and slid in and out of racks. Researchers at Purdue have been studying consumer attitudes towards sturdier biodegradable containers for flowers and plants. There has also been research in biodegradable metals, most of which has focused on medical devices. A key factor will be cost, and whether the economics of a biodegradable design for a server chassis can work at scale.
One of goals of the Open Compute Project is to separate the technology refresh cycle for CPUs and other components from the surrounding equipment, including racks and chassis. Frank Frankovsky, Facebook’s VP of Hardware Design and Supply Chain and a key driver in Open Compute, says the ability to easily swap out CPUs could transform the way chips are procured at scale, perhaps shifting to a subscription model.
Power supplies and DDR3 memory can also last through several server refresh cycles, noted Matt Corddry, Facebook’s Manager, Hardware Design, in a briefing on Facebook’s server designs earlier this year. “We believe the rack has a much longer life than some of the compute nodes,” said Corddry. “Why do we scrap all that equipment every three years? I think that’s the next frontier of efficiency in data center operations. The equipment is interchangeable.”