Distributed Computing at Multidimensional Scale
Wednesday, December 3. 9:30am - 10:30pm
by: Alfred Spector, Google (Slides of the keynote)
Abstract: One of Google's core strengths has been its distributed computing technology: The company has built a plethora of components that together form the core of a scalable system supporting a wide collection of applications operating on a global scale. This talk will reflect on the diversity of technology employed, the challenges that have been met, and the open challenges that face both us (and the greater research community) as we attempt to develop systems of even greater flexibility, resilience, capability, and efficiency.
Bio: Alfred joined Google in November of 2007 and is responsible for the research across Google and also a growing collection of special engineering initiatives - projects like Google Health.
Previously, Alfred was Vice President of Strategy and Technology IBM's Software Business, Vice President of Services and Software Research across IBM, and a General Manager in the Software Business where he led products such as CICS and WebSphere. He was also founder and CEO of Transarc Corporation, a pioneer in distributed transaction processing and wide area file systems that was acquired by IBM in 1994. Previously Alfred was Associate Professor of Computer Science at Carnegie Mellon University, specializing in highly reliable, highly scalable distributed computing.
Alfred received his Ph.D. in Computer Science from Stanford and his A.B. in Applied Mathematics from Harvard. He has published numerous articles and is a member of the National Academy of Engineering, a Fellow of the IEEE and ACM, and the recipient of the 2001 IEEE Computer Society's Tsutomu Kanai Award for work in scalable architectures and distributed systems.
2P, DSM, and Other Products of the Complexity Factory
Friday, December 5. 9:00am - 10:00pm
by: Willy Zwaenepoel, School of Computer and Communication Sciences, EPFL, Switzerland (Slides of the keynote)
Abstract: In order to get your paper accepted at a major conference, the idea you develop in the paper must be complex, preferably even incomprehensible to all but the few experts. In order to have your idea have any impact in a real system, it must be simple and comprehensible to the above-average programmer in industry. The obvious net result of this contradiction is that very few papers at major conferences have any impact in real systems. This talk will explore some examples of this dilemma, some counterexamples of ideas that were successfully transferred to practice, and some ideas on how we can perhaps improve the situation.
Bio: Willy Zwaenepoel is professor and dean of the School of Computer and Communication Sciences at EPFL in Lausanne, Switzerland. He received his B.S. from the University of Gent in 1979, and his M.S. and Ph.D. from Stanford, in 1980 and 1984, respectively. Before joining EPFL in 2002, Willy Zwaenepoel was on the faculty at Rice University, where he was the Karl F. Hasselmann Professor of Computer Science and Electrical and Computer Engineering.
He has worked in a variety of aspects of operating systems and distributed systems, including microkernels, fault tolerance, parallel scientific computing on clusters of workstations, clusters for web services, and mobile computing. He is most well known for his work on the Treadmarks distributed shared memory system, which was licensed to Intel and became the basis for Intel's OpenMP cluster product. His work on high-performance software for network I/O led to the creation of iMimic Networking, Inc, which he led from 2000 to 2005, and which is now part of Cisco Systems.