8:30 AM-9:15 AM
Room: Rio Mar 5
Chair: Richard A. Tapia, Rice University, USA
The recent report of the President's Information Technology Advisory Committee (PITAC), which was released during my term as co-chair, identified a drift away from the sort of long-term high-risk research that has delivered today's booming Internet economy. If this trend is not reversed, the flow of ideas and people into the IT industry will be threatened in the not-too-distant future. The report challenged the federal Government to increase its base of support for information technology research and shift the focus toward more fundamental topics. Three key areas of investment were specifically identified as needing increases: software, scalable information infrastructure, and high-end computing.
At the same time, the report challenged the information technology research community to rethink their research directions so that they would be ready to accept the increased emphasis on big ideas and long term projects. High performance computing has a history of fostering short-term research in software areas because of the small size of the market, which leaves it to the research community to produce software solutions at the high end. As a result, much of the research activity has been focused on the solutions of near-term problems such as the next compiler, debugger, or run-time library for direct programming of scalable machines.
If we are to accept the PITAC challenge, we need to address some truly great IT research problems for HPC Software. This talk will review the PITAC report and suggest a few promising directions for the longer term. It will also discuss the role of the SIAM community in future information technology research.
Ken Kennedy