Problem Solving Environments for Parallel Adaptive Applications

Manish Parashar & James C. Browne
{parashar,browne}@cs.utexas.edu
Department of Computer Sciences
University of Texas at Austin,
Austin, TX

Overview

This project investigates the design and development of a problem solving environment (PSE) for parallel/distributed dynamically adaptive applications. The key components of the PSE are shown in the figure below. The design builds on components developed as part of my doctoral (performance prediction and evaluation tools) and post-doctoral (data-management infrastructure) research and has resulted in the development of three domain specific PSE's listed below.

Domain Specific PSE's

  1. IPARS++: A PSE for porous media and reservoir simulations in collaboration with the Center for Subsurface Modeling, University of Texas at Austin, Austin, TX.
  2. The Framework: A PSE for interactive relativistic and hydrodynamics simulations in collaboration with Max-Plack-Institut, Potsdam, Germany, National Institute for Supercomputing Applications (NCSA), Urbana, Il, University of Valencia, Spain, and University of the Balearic Islands, Palma de Mallorca, Spain.
  3. Computational Toolkit For Numerical Relativity A computational infrastructure for binary-black hole simulations under development in collaboration with the Binary Black-Hole NSF Grand Challenge Alliance.

Related Reports and Presentations

  1. A Common Computational Infrastructure for Adaptive Algorithms for PDE Solutions
  2. Contructing a Problem Solving Environment for Porous Media Flow Computations Theatre presentation, Supercomputing '96, Pittsburg, November, 1996.
  3. A Computational Infrastructure for Parallel Adaptive Porous Media Computations ACTI Industrial Affiliates Meeting, Austin, November, 1996.
  4. Deriving a Problem Solving Environment for Parallel Adaptive Computations Invited panel presentation, ARPA-NSF Workshop on Scalable Scientific Software Libraries & Problem Solving Environments, Purdue University, September, 1995.

Manish Parashar, parashar@caip.rutgers.edu
CAIP Center & Department of Electrical and Computer Engineering
Rutgers, The State University of New Jersey