Making Enterprise Computing "Green": Energy-Efficiency Challenges in Enterprise Data Centers

Tuesday October 14, 2008
Hamerschlag Hall D-210
4:30 pm

Thomas Wenisch
University of Michigan

Enterprise data centers consume an alarmingly-high fraction of the world's energy. The total carbon footprint of the world's data centers is roughly the same as the CO2 emissions of the entire Czech Republic. In the US, the EPA estimates that data center energy consumption will reach over 100 billion kWh by 2011, 2.5% of domestic power generation (more than the nation's color televisions), resulting in an estimated annual electricity cost of $7.4 billion. Improving the energy efficiency of enterprise computing is a critical challenge for computer systems research.

In the first half of my talk, I will describe the architecture of a typical data center, introduce some of the key challenges to data center energy efficiency, and survey some promising ongoing work toward addressing these challenges. One of the largest sources of energy-inefficiency is the substantial energy used by idle equipment that is powered on, but not performing useful work. In the latter half of my talk, I will describe PowerNap, our approach to eliminate server idle-power waste. Under PowerNap, the entire system transitions rapidly between a high-performance active state and a near zero-power idle state in response to instantaneous load. Through analysis of utilization traces collected from enterprise-scale commercial deployments, I show that PowerNap can reduce average server power consumption by 76%.

Thomas Wenisch is an Assistant Professor of Computer Science and Engineering at the University of Michigan, specializing in computer architecture. Tom's prior research includes memory streaming for commercial server applications, store-wait-free multiprocessor memory systems and rigorous sampling-based performance evaluation methodologies.

He is a principle developer of the Flexus full-system cycle-accurate simulation infrastructure. His ongoing work focuses on data center architecture, energy-efficient server design, and multi-core / multiprocessor memory systems. Prior to his academic career, Tom was a software developer at American Power Conversion, where he worked on data center thermal topology estimation. He is co-inventor on three patents. Tom received his Ph.D. in Electrical and Computer Engineering from Carnegie Mellon University.