A-MPI : supporting MPI on a nondedicated cluster of workstations
Abstract (Summary)
Distributing data is one of the fundamental problems in implementing efficient
distributed-memory parallel programs. The problem becomes more difficult in environments
where the participating nodes (processors) are not dedicated to a parallel
application. Such environments increase the difficulty of the data distribution
problem, which is to determine an assignment of data elements to each node to minimize
completion time. We are investigating this problem in the context of explicit
message-passing programs.
We have designed and implemented an extension to the popular Message Passing
Interface (MPI) that efficiently supports adaptive programs by providing the necessary
infrastructure to redistribute data dynamically. Our system, called the Adaptive
Message Passing Interface (A-MPI), contributes (1) an efficient memory allocation
mechanism, (2) techniques for accurately determining system load and computation
time, and (3) a heuristic for determining efficient data distributions, including the
removal of nodes whose participation degrades the performance of an application.
Performance results show that programs that use A-MPI can produce significant
improvements over previous load-balancing systems.
Index words: Load Balancing, MPI, Data Distribution, Parallel
Programming
A-MPI: Supporting MPI on a Nondedicated Cluster of Workstations
by
D. Brent Weatherly
B.S., The University of Georgia, 1994
A.B., The University of Georgia, 1995
A Thesis Submitted to the Graduate Faculty
of The University of Georgia in Partial Fulfillment
of the
Requirements for the Degree
Master of Science
Athens, Georgia
2002
c? 2002
D. Brent Weatherly
All Rights Reserved
A-MPI: Supporting MPI on a Nondedicated Cluster of Workstations
by
D. Brent Weatherly
Approved:
Major Professor: David Lowenthal
Committee: John Miller
Eileen Kraemer
Electronic Version Approved:
Gordhan L. Patel
Dean of the Graduate School
The University of Georgia
August 2002
Bibliographical Information:
Advisor:
School:The University of Georgia
School Location:USA - Georgia
Source Type:Master's Thesis
Keywords:
ISBN:
Date of Publication: