A pilot study to compare programming effort for two parallel programming models

Writing software for the current generation of parallel systems requires significant programmer effort, and the community is seeking alternatives that reduce effort while still achieving good performance. Measure the effect of parallel programming models (message-passing vs. PRAM-like) on programmer...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of systems and software Jg. 81; H. 11; S. 1920 - 1930
Hauptverfasser: Hochstein, Lorin, Basili, Victor R., Vishkin, Uzi, Gilbert, John
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Elsevier Inc 01.11.2008
Elsevier Sequoia S.A
Schlagworte:
Comparative analysis
Computer programming
Effort
Empirical study
Message-passing
MPI
Parallel programming
PRAM
Problem solving
Programmers
Studies
XMT
Parallel programming
Empirical study
MPI
Message-passing
Effort
PRAM
XMT
ISSN:0164-1212, 1873-1228
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Writing software for the current generation of parallel systems requires significant programmer effort, and the community is seeking alternatives that reduce effort while still achieving good performance. Measure the effect of parallel programming models (message-passing vs. PRAM-like) on programmer effort. One group of subjects implemented sparse-matrix dense-vector multiplication using message-passing (MPI), and a second group solved the same problem using a PRAM-like model (XMTC). The subjects were students in two graduate-level classes: one class was taught MPI and the other was taught XMTC. Development time, program correctness. Mean XMTC development time was 4.8 h less than mean MPI development time (95% confidence interval, 2.0–7.7), a 46% reduction. XMTC programs were more likely to be correct, but the difference in correctness rates was not statistically significant ( p = .16). XMTC solutions for this particular problem required less effort than MPI equivalents, but further studies are necessary which examine different types of problems and different levels of programmer experience.
Bibliographie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ISSN:0164-1212
1873-1228
DOI:10.1016/j.jss.2007.12.798