2011 Poster Sessions : Towards Energy Efficient Computing in the Enterprise

Student Name : Maria Kazandjieva
Advisor : Philip Levis
Research Areas: Computer Systems
Abstract:
PowerNet is a hybrid sensor network for monitoring the power and utilization of computing systems in a large academic building. PowerNet comprises approximately 300 single-plug wired and wireless hardware power meters and 23 software sensors that monitor PCs, laptops, network switches, servers, LCD screens, and other office equipment.

This dense, long-term monitoring allows us to extrapolate the energy< consumption breakdown of a whole enterprise building. Using our measurements together with device inventory we find that approximately 56% of the total building energy budget goes toward computing systems, at a cost of ~$22,000 per month. PowerNet's measurements of CPU activity and network traffic reveal that a large fraction of this power is wasted and shows where there are savings opportunities.

Over the last few years, researchers have proposed a variety of enterprise-focused energy-saving techniques, from network proxies and virtual machine migration to the return of thin clients. Using our long term measurements, we argue that software solutions should come second, if at all, only after hardware efficiency has been addressed. Specifically, we show that smart choices of computing equipment, in the form of laptops, can have a higher impact on energy conservation than state-of-the-art power management software.

Bio:
Maria is a fourth-year Computer Science PhD student at Stanford University. Maria is part of the SING lab that tackles wireless networking and systems research questions. Her research interests lie in sensor networks, green computing, and data visualization. Before coming to Stanford, Maria earned a BA in Computer Science at Mount Holyoke College and worked at Princeton University as research staff.