Seminar by Chandu Thekkath
Enabling Large-Scale Data Intensive Computations
Chandu Thekkath
Microsoft Research
Date: Monday, September 6th, 2010
Time: 4:00 PM
Venue: CS102.Abstract:
This talk describes a set of distributed services developed at Microsoft Research Silicon Valley to enable efficient parallel programming on very large datasets. Parallel programs arise naturally within scientific, data mining, and business applications. Central to our philosophy is the notion that parallel programs do not have to be difficult to write and that the same program must seamlessly run on a laptop, desktop, a small cluster, or on a large data center without the author having to worry about the details of parallelization, synchronization, or fault-tolerance. Dryad and DryadLINQ are two services that embody this belief. The combination is extensively used within Microsoft, and is available free to academics, researchers, and non-commercial users. Our goal is to enable users, particularly non computer-scientists, to treat a computer cluster as a forensic, diagnostic, or analytic tool. The talk will describe the details of the system and the characteristics of some of the applications that have been run on it.
About the speaker:
Chandu Thekkath has returned to the ranks of MSR Silicon Valley after a two-year stint as the Director of the Platforms and Distributed Systems Group in within Microsoft Research. Before that he was a Principal Researcher at Microsoft Research in Silicon Valley. In that position, he worked closely with the Hotmail team on the Blue project, before which he worked on the Boxwood and Koh-i-Noor projects. He worked as the interim head of the Hardware, Communications, and Systems group in Microsoft Research, Bangalore during 2006. He joined Microsoft in 2001 from the DEC/Compaq Systems Research Center, where he held the positions of Principal Engineer, Consulting Engineer, and Manager (Distributed Systems).
Thekkath's research expertise spans operating systems, distributed systems, and networks. He holds several patents in operating systems, networks, distributed systems, and computer architecture. He has published a variety of papers in the premier conferences on experimental computers systems such as SOSP, ASPLOS, SIGCOMM, and OSDI. He has also served on the programme committees of SOSP, ASPLOS, OSDI, NSDI, and FAST, which he also chaired in 2004. He was made a fellow of the ACM for his contributions to operating systems, distributed systems, and scalable storage.
Thekkath received a BTech. in EE (Electronics) from IIT Madras in 1982, an M.S. in EE from UC Santa Barbara in 1983, an M.S. in Computer Science from Stanford in 1989, and a Ph.D. in Computer Science from the University of Washington in 1994. He worked as a software development engineer at Monolithic Memories Inc. (now part of AMD) and Hewlett Packard between 1983 and 1988.