Invited Speakers

Dynamic Power Management in Data Centers: Theory & Practice
Mor Harchol-Balter
, Professor, Computer Science Department, Carnegie Mellon University

Energy costs for data centers continue to rise, already exceeding ten billion dollars yearly. Sadly much of this power is wasted. Servers are only busy 10-30% of the time, but they are often left on, while idle, utilizing 60% of more of peak power while in the idle state. The obvious solution is dynamic power management: turning servers off, or re-purposing them, when idle. The drawback is a prohibitive "setup cost" to get servers back on. The purpose of this talk is to understand the effect of the "setup cost" and whether dynamic power management makes sense.

We first turn to theory and study the effect of setup cost in an M/M/k queue. We present the first analysis of the M/M/k/setup queueing system. We do this by introducing a new technique for analyzing infinite, repeating, Markov chains, which we call Recursive Renewal Reward (RRR).

We then turn to implementation, where we implement and evaluate dynamic power management in a multi-tier data center with key-value store workload, reminiscent of Facebook or Amazon. We propose a new dynamic algorithm, AutoScale, which is ideally suited to the case of unpredictable, time-varying load, and we show that AutoScale dramatically reduces power in data centers.

Joint work with: Anshul Gandhi, Alan Scheller-Wolf, and Mike Kozuch.

Mor Harchol-Balter is a Professor of Computer Science at Carnegie Mellon University. From 2008-2011, she served as the Associate Department Head for Computer Science. Mor received her doctorate from the Computer Science department at U.C. Berkeley under the direction of Manuel Blum in 1996 and then spent three years at MIT under the NSF Postdoctoral Fellowship in the Mathematical Sciences, before joining CMU. She is a recipient of the McCandless Chair, the NSF CAREER award, multiple best paper awards, and several teaching awards, including the Herbert A. Simon Award for Teaching Excellence. Mor is heavily involved in the ACM SIGMETRICS research community, where she served as Technical Program Chair for Sigmetrics 2007 and as General Chair for Sigmetrics 2013. Mor's work focuses on designing new resource allocation policies (load balancing policies, power management policies, and scheduling policies) for server farms. In 2013, Mor authored her first textbook, "Performance Analysis and Design of Computer Systems," published by Cambridge University Press, 2013. Mor is known for both her work in queueing analysis and in systems implementation, but she is perhaps best known for her many successful PhD students, the majority of whom are professors at top universities.

Sustainability as a Path for ICT Resiliency to Natural Disasters
Alexis Kwasinski
, Associate Professor, Electrical & Computer Engineering, University of Texas at Austin

This talk explores the concepts of sustainability and resiliency within the context of ICT systems performance during natural disasters. In recent past natural disasters, ICT systems performance had important issues that in some cases led to service disruptions. One of the most significant cause of service disruptions in ICT systems during these recent natural disasters originated in electric power supply issues. Hence, this talk will discuss how more sustainable electric power systems, including smart grid technologies and microgrids, may create a more resilient support infrastructure for ICT systems and improve their performance during natural disasters. The discussion will be based on practical perspectives gained during field damage assessments after notable recent natural disasters that had a significant impact on ICT systems, including hurricanes Katrina, Ike and Sandy and the 2011 earthquake and tsunami in Japan.

Alexis Kwasinski has an extensive academic and industry experience in the ICT power and infrastructure areas. He earned his M.S. and Ph.D. degrees in electrical engineering from the University of Illinois at Urbana-Champaign (UIUC) in 2005 and 2007, respectively. Previously, he spent almost 10 years working for Telefónica of Argentina and for Lucent Technologies Power Systems. He is currently an Associate Professor in the Department of Electrical and Computer Engineering at The University of Texas at Austin and his research interests include power electronic systems, distributed generation (microgrids), renewable and alternative energy, smart grids, and analysis of the impact of natural disasters on critical power infrastructure. He participated in field damage assessments after natural disasters, including hurricane Katrina and the March 2011 earthquake and tsunami in Japan. In 2005, Dr. Kwasinski was awarded the Joseph J. Suozzi INTELEC Fellowship and in 2007 he received the best technical paper award at INTELEC. In 2009 he received an NSF CAREER award and in 2011 he received an IBM Faculty Innovation Award. Dr. Kwasinski is also an Associate Editor for the IEEE Transactions on Energy Conversion and IEEE Transactions on Power Electronics.

Challenges to Assessing and Providing Energy Efficient Computing
Karthick Rajamani
, IBM Research

In this talk, I will share different challenges I came across while developing and advocating the adoption of energy efficient technologies in servers and data centers. Specifically, I will talk about challenges in measuring power and assessing energy efficiency of servers, and challenges posed by real and imagined trade-offs between efficiency and performance/reliability. The audience will hopefully gain healthy appreciation for some not-so-obvious challenges in bringing new power management technologies into products and for more holistic approaches to energy efficiency.

Karthick Rajamani is a Research Staff Member in IBM Research Austin and manager of the Computing-As-a-Service-Technologies (CAST) department. As a member and later manager of the Power-aware Systems department, he made key contributions to different dynamic power management architectures from embedded systems to high-end servers including designs of the IBM POWER6, POWER7 and POWER8 processors and systems and the IBM EnergyScale technology. His current research includes cloud computing technologies and platforms, emerging memory technologies, and energy efficient, reliable computing. He graduated with a B.Tech. degree in Electronics and Communications from Indian Institute of Technology, Madras and obtained his Master’s and Ph.D. degrees in Electrical and Computer Engineering, from Rice University, Houston.