DocumentCode :
2406579
Title :
MOPPS: a scheme for managing parallel scientific programs in a distributed architecture
Author :
Arafeh, Hassan ; Alkhatib, Hasan S. ; Barraclough, Hal T.
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Santa Clara Univ., CA, USA
fYear :
1990
fDate :
Feb. 26 1990-March 2 1990
Firstpage :
387
Lastpage :
394
Abstract :
MOPPS, a system under development at Santa Clara University, which employs a new solution to the problem of managing programs which solve scientific and engineering applications on a distributed computer environment, is introduced. The core of MOPPS is an intelligent program manager that builds a database of the execution performance of parallel programs it manages under different conditions. The manager then applies that knowledge to improve the performance of future runs. The program manager learns from experience. MOPPs is not locked into a specific architecture, programming language, or operating system. It looks at a program as a collection of functions that do specific scientific programming tasks.<>
Keywords :
distributed processing; expert systems; parallel programming; supervisory programs; MOPPS; distributed architecture; execution performance; intelligent program manager; managing programs; parallel programs; parallel scientific programs; Application software; Computer architecture; Distributed computing; Engineering management; Environmental management; Hardware; Job shop scheduling; Network topology; Operating systems; Processor scheduling;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Compcon Spring '90. Intellectual Leverage. Digest of Papers. Thirty-Fifth IEEE Computer Society International Conference.
Conference_Location :
San Francisco, CA, USA
Print_ISBN :
0-8186-2028-5
Type :
conf
DOI :
10.1109/CMPCON.1990.63713
Filename :
63713
Link To Document :
بازگشت