Stanford Networking Seminar

12:15PM, Thursday February 25, 2010
Gates 104

A Data Center Congestion Control Algorithm: The IEEE 802.1Qau Standard

Balaji Prabhakar
Stanford University

About the talk:
Data Center Networks present a novel, unique and rich environment for algorithm and protocol development. Projects are underway at the IEEE 802.1 standards body, especially in the Data Center Bridging Task Group, to define new Ethernet functions for data center use. We describe one such project, IEEE 802.1Qau, which aims to develop an Ethernet congestion control algorithm. An analysis of the algorithm (called QCN) developed for the standard led to the discovery of a simple method for making congestion control loops stable in the face of increasing lags. This method, called the Averaging Principle, is applicable for stabilizing general control loops, not just congestion control loops. We describe the development of the QCN algorithm and the Averaging Principle.

About the speaker:
Balaji Prabhakar is an Associate Professor in the Departments of Electrical Engineering and Computer Science at Stanford University. His research interests are in computer networks; notably, in designing algorithms for the Internet and for Data Centers. Recently, he has been interested in Societal Networks: networks vital for society's functioning, such as transportation, electricity and recycling systems. He has been involved in developing and deploying incentive mechanisms to move commuters to off-peak times so that congestion, fuel and pollution costs are reduced. He has been a Terman Fellow at Stanford University and a Fellow of the Alfred P. Sloan Foundation. He has received the CAREER award from the U.S. National Science Foundation, the Erlang Prize, the Rollo Davidson Prize, and delivered the Lunteren Lectures in Holland. He is a co-recipient of best paper awards at Hot Interconnects 2002, IEEE Infocom 2004 and ACM Sigmetrics 2008.