Storage for Extreme-Scale Computing
Dr. Rob Ross
Computer Scientist, Math & Computer Science Div, Argonne National Lab.
Senior Fellow, Computation Institute, University of Chicago
Stuart Building 223
Tuesday, November 1st, 2011
11:00AM - 12:00PM
Abstract: Storage systems play a crucial role in high-performance computing, providing a fault tolerant repository for application state and holding the results of simulations for future analysis. Without robust storage systems, high-performance computing applications would fail to deliver new discoveries. Supercomputing systems will reach unprecedented levels of concurrency, complexity, and rate of computation in the near future, dramatically widening the gap between simulation and storage capabilities. To meet the needs of extreme- scale applications, future storage systems must break from business as usual. In this talk we will discuss some of the drivers for research in the area of storage for extreme-scale systems and highlight four areas in need of attention in order for storage systems to succeed over the coming decade.
Bio: Robert Ross is a computer scientist at Argonne National Laboratory and a senior fellow at the Computation Institute of the University of Chicago. Rob's research interests are in system software for high performance computing systems, in particular parallel file systems and libraries for I/O and message passing. Rob received his Ph.D. in Computer Engineering from Clemson University in 2000. Following this he joined the Mathematics and Computer Science Division at Argonne National Laboratory. He was the lead architect of the PVFS parallel file system, is active in maintaining the ROMIO MPI-IO library, and oversees the Parallel netCDF high-level I/O library development. Rob is a member of the MPICH2 development team awarded the R&D 100 award in 2005, and was a recipient of the 2004 Presidential Early Career Award for Scientists and Engineers.