2007 SOS11

2005 SOS

The Meeting

One of the SOS8 program's principal goals is to assemble interdisciplinary teams and collaboratories to develop the necessary state-of-the-art mathematical algorithms and software, supported by appropriate hardware and middleware infrastructure, to use terascale computers effectively to advance fundamental research in science central to the SOS8 science mission. The overarching goal of the first SOS meeting was to familiarize all participants with the overall integrated SOS program and to initiate team building, both across DOE laboratories and with academia to support that integration across projects.

The theme this year is "Advanced Computer Architectures for Science."

We will be discussing machines such as Sandia's Redstorm, Cray's X1, IBM's BlueGene, and any other architectures being designed to solve "ultra-scale science problems."

Discussion Topics:

  • What are the computational needs of particular science areas (not just power, but I/O, memory per node, disk, etc.)
  • What are the characteristics of these machines that make them suitable for science.
  • On what types of applications will they run well, where will they run poorly?
  • How can we work together as a community to avoid scaling bottlenecks in the hardware and software so that scientists can reach ultra-scale (100+ TF) sustained performance in the next 3 years?

Ultra-scale Science areas to be addressed include: nanoscience, biology, climate, and chemistry. Others?

Charleston historic district

disclaimer    search