Open Access
Towards a Portable Parallel Programming Environment
Rebecca Harris,Guang R. Gao,Herbert H. J. Hum,Julian Lebensold,Laurie Hendren,Lixin Tao,Mike Feeley,Marius V. A. Hancu,Prakash Panangaden,Vincent Van Dongen +9 more
- 01 Jan 1992
7
TL;DR: The EPPP environment will be portable in the sense that the user will be able to rapidly port his/her application on a variety of parallel architectures, and integrates a parallel programming paradigm, a compiling methodology, a simulator, and a complete set of analysis tools.
read more
Abstract: This paper presents an important project whose objective is to develop a Portable Parallel Programming Environment, called EPPP, for current and future generation parallel computers. Our environment will be portable in the sense that the user will be able to rapidly port his/her application on a variety of parallel architectures. That is, our goal is to allow the programmer to develop, debug and tune his/her application on a workstation; a simulator will provide the user's execution platform at low cost. To produce an executable program for a particular architecture, the high-level program will be combined with domain-speciic and architecture-speciic information to produce a compiled program that will run eeciently on the speciied architecture. This information will be provided by the environment, the user, or a combination of both. The proposed environment integrates a parallel programming paradigm, a compiling methodology, a simulator, and a complete set of analysis tools. Benchmark programs will include the traditional scientiic array-based applications and more general-purpose applications that use dynamic data structures.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Automatic data and computation decomposition for distributed memory machines
Qi Ning,Vincent Van Dongen,Guang R. Gao +2 more
- 04 Jan 1995
TL;DR: An automatic compile-time computation and data decomposition technique for distributed memory machines that can handle complex programs containing perfect and nonperfect loop nests with or without loop-carried dependences and can selectively relax data locality constraints within a cluster to achieve a balance between parallelism and data locality.
A Performance Debugger for a Language Supporting Data Distribution Primitives
G. Hurteau,Ajit Singh,M. Hancu,V. Van Dongen,Herbert H. J. Hum +4 more
- 15 Aug 1994
TL;DR: The design and implementation of a performance debugger that is specifically targeted to meet the performance debugging requirements of a data-parallel programming model based on user-specified data distributions is presented.
4
•Proceedings Article
Automatic decomposition in EPPP compiler
Qi Ning,Vincent Van Dongen,Guang R. Gao +2 more
- 31 Oct 1994
TL;DR: An overview of the automatic computation and data decomposition scheme in the EPPP compiler for distributed memory machines, which can also be used in compilers for programming languages such as HPF in conjunction with the user-specified data distribution directives.
1
Experiences with Parallel Programming in Meteorology
Fritz G. Wollenweber
- 01 Jan 1995
TL;DR: The efforts described in this paper are the work of a number of people and special appreciation is due to the German collaborators Ute Gaertel, Toni Schueller and Wolfgang Joppisch, who did the major part of actually programming the parallel version of IFS available at that time.
References
•Book
Supercompilers for parallel and vector computers
Hans P. Zima,Barbara Chapman +1 more
- 01 Jan 1990
TL;DR: This paper presents a meta-modelling architecture for supercompilers that automates the very labor-intensive and therefore time-heavy and expensive process of learning and optimization of supercomputing systems.
778
•Proceedings Article
Parallelizing Programs with Recursive Data Structures.
Laurie Hendren,Alexandru Nicolau +1 more
- 01 Jan 1989
TL;DR: In this paper, the authors present interference analysis tools and parallelization techniques for imperative programs that contain dynamically updatable trees and directed acyclic graphs, based on a regular-expression-like representation of the relationship between accessible nodes in the data structure.
131
Interprocedural aliasing in the presence of pointers
William Landi
- 01 Jan 1992
TL;DR: An algorithm which safely approximates Interprocedural May Alias (i.e., the authors fail to report an alias only if it never occurs) in the presence of pointers is presented and it is shown that the algorithm is as precise as possible in the worst case.
89
Chief: a parallel simulation environment for parallel systems
J.D. Bruner,Hoichi Cheong,Alexander V. Veidenbaum,Pen-Chung Yew +3 more
- 30 Apr 1991
TL;DR: The Chief project provides a environment for analyzing parallel systems and aims to employ the Perfect club benchmarks to evaluate performance.
19
Mapping parallel programs onto parallel systems with torus and mesh-based communication structures
Lixin Tao,Eva Ma +1 more
- 01 Jan 1988
TL;DR: This research introduces a rich class of low dilation cost graph embedding functions for toruses and meshes of various dimensions and various shapes and proposes an efficient program loading approach based on inverses of mapping functions and a broadcast network.
5
Related Papers (5)
Wolfgang K. Giloi
- 01 Jan 1992
Kai Hwang
- 01 Jun 1993
B. Ramkumar,Prithviraj Banerjee +1 more
- 11 Oct 1992
Manohar Rao,Zary Segall,D. Vrsalovic +2 more
- 01 Oct 1990