Abstract:
Today"s capability-class massively parallel processing (MPP) machines have to tens-to-hundreds of thousands of processors, with next-generation systems planned to have in excess of one million processors. For systems of such scale, I/O is a significant challenge that may not be solved using traditional approaches. In this talk, I will describe some of the challenges facing I/O for the next generation of HPC systems and present research at Sandia National Laboratories to develop "I/O Services" that leverage available memory and processing within the HPC system to offset the I/O imbalance inherent in these systems, and to provide new capabilities to HPC applications.

Biography:
Ron A. Oldfield is a senior member of the technical staff at Sandia National Laboratories in Albuquerque, NM. He received the B.Sc. in computer science from the University of New Mexico in 1993. From 1993 to 1997, he worked in the computational sciences department of Sandia National Laboratories, where he specialized in seismic research and parallel I/O. He was the primary developer for the GONII-SSD (Gas and Oil National Information Infrastructure--Synthetic Seismic Dataset) project and a co-developer for the R&D 100 award winning project "Salvo", a project to develop a 3D finite-difference prestack-depth migration algorithm for massively parallel architectures. From 1997 to 2003 he attended graduate school at Dartmouth college and received his Ph.D. in June, 2003. In September of 2003, he returned to Sandia. He currently leads a number of I/O, resilience, and informatics projects and is project manager of over 10 active projects in the areas of scalable I/O and resilience. His research interests include parallel and distributed computing, parallel I/O, resilience, and performance modeling.