Problem Specific Environments for Parallel Systems.
Abstract
Parallelism is one of the key components of large scale, high performance computing. Extensive use of parallelism has yielded a tremendous increase in the raw processing speed of high performance systems, but parallel problem solving remains difficult. These difficulties are typically solved by building software tools, such as parallel programming environments. Existing parallel programming environments are general purpose, using a broad paradigm. This report illustrates that problem specific environments are more beneficial than general purpose environments by making a case study comparison of two classes of problems. The simple class consists of two classic graph problems, namely, all pairs shortest path and connected components. The complex class consists of an elliptic partial differential equations solved via domain decomposition. The general purpose tools BUILD and SCHEDULE are used to solve each class. A problem specific environment for solving each class is then described. Comparisons between special purpose environments and general purpose environments show that the special purpose environments allow the user to concentrate on the problem, while general purpose environments force the user to think about mapping the problem to the environment rather than solving the problem in parallel.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1993
- Accession Number
- ADA266129
Entities
People
- Loretta S. Auvil
Organizations
- Rome Laboratory