Exploring the Efficiency of Renewable Energy-based Modular Data Centers at Scale

Abstract: Modular data centers that can be placed right at the energy farms and powered mostly by renewable energy, provide a flexible and effective approach to lowering the carbon footprint of data centers. However, the main challenge of using renewable energy is the high variability of power produced, which implies large volatility in powering computing resources at modular data centers, and degraded application performance due to the task evictions and migrations. In this talk, we will show that, with careful deployment of modular data centers, we can have the lowest carbon emissions in data centers, while minimizing the impact of the power variability on cloud virtual machines.

Bio: Jian Huang is an Associate Professor and Y. T. Lo Faculty Fellow in the ECE department at UIUC. His research interests include computer systems and architecture, AI infrastructure, memory/storage systems, systems security, and especially the intersections of them. Most recently, he has been working on sustainable AI infrastructures. His research contributions have been published at computer architecture and systems conferences such as ISCA, MICRO, ASPLOS, OSDI, and SOSP. He received the inaugural SIGMICRO Early Career Award, NSF CAREER Award, NSF CRII Award, USENIX Best Paper Award, MICRO Top Picks, MICRO Hall of Fame, Dean’s Award for Early Innovation, and a few industry awards. He is also the founder of the workshop on Hot Topics in System Infrastructure (HotInfra).