Exascale Data Analysis

Traditionaly, the primary role of supercomputers was to create data, primarily for simulation applications. Due to usage and technology trends, supercomputers are increasingly also used for data analysis. Some of this data is from simulations, but there is also a rapidly increasingly amount of real-world science and business data to be analyzed. We briefly overview Blue Gene and other current supercomputer architectures. We outline future architectures, up to the Exascale supercomputers expected in the 2020 time frame. We focus on the data analysis challenges and opportunites, especially those concerning Flash and other up-and-coming storage class memory. About the speakers Blake G. Fitch has been with IBM Research, Yorktown Heights, NY since 1987, mainly pursuing interests in parallel systems. He joined the Scalable Parallel Systems Group in 1990, contributing to research and development that culminated in the IBM scalable parallel system (SP*) product. His research interests have focused on application frameworks and programming models suitable for production parallel computing environments. Practical application of this work includes contributions to the transputerbased control system for the 1994 IBM CMOS S/390* mainframes, while at IBM Boeblingen, Germany, and to the architecture of the 1996 IBM Automatic Fingerprint Identification System parallel application, while at IBM Hursley, UK. He joined the Blue Gene project in 1999 as the application architect for Blue Matter, a scalable molecular dynamics package. From 2008 he is the technical lead for data storage and analysis research in the Blue Gene program. Burkhard Steinmacher-Burow was a member of the ZEUS collaboration at DESY from 1988 to 1996. His PhD thesis measured the total photon-proton cross section. In the early 1990's he led the development of Funnel, the ZEUS production facilty for simulated events, using hundreds of workstations around the world. He has been with the Blue Gene project since 2001, primarily contributing to the network across nodes and to the memory system within a node. Initially based at IBM Research, Yorktown Heights, NY, in 2004 he moved to IBM Research and Development Germany in Boeblingen, where he also is the technical lead of the local Blue Gene team.