COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring. |
University of Cambridge > Talks.cam > Computer Laboratory Systems Research Group Seminar > Optimal-Performance Next-Generation (Datacenter) Networks
Optimal-Performance Next-Generation (Datacenter) NetworksAdd to your list(s) Download to your calendar using vCal
If you have a question about this talk, please contact Liang Wang. The rapid growth of Internet services is placing tremendous demands on the underlying networks. Yet, today’s core Internet protocols (TCP/IP, BGP , etc.) have remained fairly stagnant for decades and, consequently, often fail to cope with the ever-growing challenges. Moreover, even carefully-engineered networks, such as datacenters, typically exhibit far-from-optimal performance. This calls for the design of new and improved network protocols and architectures. I will present two recent research directions along these lines: Even carefully-engineered specialized TCP variants can be over 10x away from optimal performance. We present Performance-oriented Congestion Control (PCC), a new congestion control protocol that achieves consistently high performance even under challenging conditions. We leverage graph-theoretic insights to design a novel datacenter architecture, called Xpander. We show that Xpanders significantly outperform traditional datacenters in terms of network throughput, resiliency to failures, cost efficiency, incremental growth, and more. Joint work with Mo Dong, P. Brighten Godfrey, Qingxi Li, and Doron Zarchy (in NSDI 2015 ) and with Michael Dinitz and Asaf Valadarsky (in HotNets 2015) Bio: Michael Schapira is an associate professor at the School of Computer Science and Engineering, the Hebrew University of Jerusalem, and the scientific co-leader of the Fraunhofer Cybersecurity Center at Hebrew University. His research focuses on the design and analysis of novel (Inter)network architectures and protocols. Prior to joining the Hebrew University, he was at Google NYC ’s Infrastructure Networking Group and a postdoctoral researcher at UC Berkeley, Yale University, and Princeton University. He is a recipient of the Allon Fellowship, the Microsoft Research Faculty Fellowship, the IETF /IRTF Applied Networking Research Prize, the Hebrew University President’s Prize, and the Krill Prize. He holds a B.Sc. in Mathematics and Computer Science, a B.A. in Humanities, and a Ph.D. in Computer Science, all from the Hebrew University (awarded in 2004, 2004, and 2008, respectively). This talk is part of the Computer Laboratory Systems Research Group Seminar series. This talk is included in these lists:
Note that ex-directory lists are not shown. |
Other listsMachine Learning Sport Looking at Language Acquisition (LALA) XIII - A meeting of Essex and Cambridge PhD students DAMTP Jubilee Celebration Economic Epidemiology 12th Annual Disability LectureOther talksLanguage Adaptation experiments: Cross-lingual embeddings for related languages Not 'just a GP' Babraham Lecture - The Remote Control of Gene Expression Mesembs - Actual and Digital Aspects of adaptive Galerkin FE for stochastic direct and inverse problems |