Scientific Computing Research Environments for the Mathematical Sciences (SCREMS)

University Of Pittsburgh, Pittsburgh PA

Investigators

Abstract

PROJECT SUMMARY The Department of Mathematics at University of Pittsburgh proposes to build a medium to large-scale scientific computing environment for research in the mathematical sciences. The computer equipment to be purchased includes an Intel 8 processor shared memory computational server, a four node Intel cluster and an Intel file server, all running Linux. This computing environment will enhance ongoing research in several ways. The availability of a local infrastructure for the investigation of shared and distributed memory techniques will result in a significant reduction in the time needed for developing software. The currently available computational power would be substantially expanded allowing for increased size and complexity in test problems and simulations. The investigators will be able to share a complete, constantly updated state-of-the-art computational science environment. The computing equipment will support several research projects, including in particular: computational methods for fluid dynamics, numerical modeling of flow in porous media, numerical methods for problems with complementarity constraints, and computational neuroscience.