Reliable Data Movement Framework for Distributed Petascale Science

Rajkumar Kettimuthu
Seminar

Abstract:
Today the science environment is very different from that of a few years ago. Large-scale collaborative science is becoming increasingly common, involving thousands of collaborators distributed across national laboratories and universities and generating massive amounts of data that must be transferred across wide area networks. Simulation sciences such as climate, computational chemistry, fusion, and astrophysics, as well as experimental sciences such as high-energy physics, nuclear physics, light sources, and fusion, are poised to produce enormous quantities of data. However, this data is useful only if it can be effectively accessed and analyzed - tasks that are challenging not only because of the sheer size of the data but also because of the distributed environment in which data is stored and analysis occurs. Critically needed is coordination of numerous shared resources, including CPUs, storage, and networks. In this talk, I will provide an overview of an extensible framework that we have developed for reliable, high-performance, and secure data movement within a distributed science environment. I will also discuss our ongoing and future research directions that focus on managing and reserving the space, bandwidth, connections, and other resources needed to transfer data to and from a storage system and thus provide a better than best effort data movement service.

Short Bio:
Rajkumar Kettimuthu is a researcher in the Mathematics and Computer Science Division at Argonne National Laboratory, USA. He is also a research staff at the University of Chicago's Computation Institute. He is the technology coordinator for Globus GridFTP and Reliable File Transfer (RFT) service. His research interests include Data Transport and Management in the Grids, and Scheduling and Resource Management for Parallel and Distributed Systems. He has published over 30 articles in these areas. He has served as program chair or co-chair for multiple international workshops. He has served on the program committee for more than 30 international conferences and workshops. He serves as a principal investigator for multiple projects on high speed data movement in distributed environments. Currently, he is listed in Who's Who in America and Who's Who in Science and Engineering.