Scheduling task graphs containing branches on parallel systems

H. El-Rewini, H. H. Ali

Research output: Contribution to journalConference article

2 Citations (Scopus)

Abstract

The problem of scheduling task graphs with conditional branching is considered one of the most difficult problems in scheduling parallel programs on multiprocessor computers. The major problem in having branches in task graphs is the non-determinism, since the direction of a branch may be unknown until the program is midway in execution. The authors overcome the problem of non-determinism by proposing a new probabilistic model that distinguishes between branch and precedence relations in parallel programs. They present two different approaches for solving this problem. In the first approach, a schedule is obtained by merging the schedules generated for several possible task graphs. The second technique generates a schedule for a single deterministic task graph that approximates all possible task graphs.

Original languageEnglish (US)
Article number184007
Pages (from-to)448-457
Number of pages10
JournalProceedings of the Annual Hawaii International Conference on System Sciences
Volume2
DOIs
StatePublished - Jan 1 1991
Event24th Annual Hawaii International Conference on System Sciences, HICSS 1991 - Kauai, United States
Duration: Jan 8 1991Jan 11 1991

Fingerprint

Scheduling
Merging
Statistical Models

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Scheduling task graphs containing branches on parallel systems. / El-Rewini, H.; Ali, H. H.

In: Proceedings of the Annual Hawaii International Conference on System Sciences, Vol. 2, 184007, 01.01.1991, p. 448-457.

Research output: Contribution to journalConference article

@article{669e3c01c0344185a171cb567b6ba8a2,
title = "Scheduling task graphs containing branches on parallel systems",
abstract = "The problem of scheduling task graphs with conditional branching is considered one of the most difficult problems in scheduling parallel programs on multiprocessor computers. The major problem in having branches in task graphs is the non-determinism, since the direction of a branch may be unknown until the program is midway in execution. The authors overcome the problem of non-determinism by proposing a new probabilistic model that distinguishes between branch and precedence relations in parallel programs. They present two different approaches for solving this problem. In the first approach, a schedule is obtained by merging the schedules generated for several possible task graphs. The second technique generates a schedule for a single deterministic task graph that approximates all possible task graphs.",
author = "H. El-Rewini and Ali, {H. H.}",
year = "1991",
month = "1",
day = "1",
doi = "10.1109/HICSS.1991.184007",
language = "English (US)",
volume = "2",
pages = "448--457",
journal = "Proceedings of the Annual Hawaii International Conference on System Sciences",
issn = "1530-1605",

}

TY - JOUR

T1 - Scheduling task graphs containing branches on parallel systems

AU - El-Rewini, H.

AU - Ali, H. H.

PY - 1991/1/1

Y1 - 1991/1/1

N2 - The problem of scheduling task graphs with conditional branching is considered one of the most difficult problems in scheduling parallel programs on multiprocessor computers. The major problem in having branches in task graphs is the non-determinism, since the direction of a branch may be unknown until the program is midway in execution. The authors overcome the problem of non-determinism by proposing a new probabilistic model that distinguishes between branch and precedence relations in parallel programs. They present two different approaches for solving this problem. In the first approach, a schedule is obtained by merging the schedules generated for several possible task graphs. The second technique generates a schedule for a single deterministic task graph that approximates all possible task graphs.

AB - The problem of scheduling task graphs with conditional branching is considered one of the most difficult problems in scheduling parallel programs on multiprocessor computers. The major problem in having branches in task graphs is the non-determinism, since the direction of a branch may be unknown until the program is midway in execution. The authors overcome the problem of non-determinism by proposing a new probabilistic model that distinguishes between branch and precedence relations in parallel programs. They present two different approaches for solving this problem. In the first approach, a schedule is obtained by merging the schedules generated for several possible task graphs. The second technique generates a schedule for a single deterministic task graph that approximates all possible task graphs.

UR - http://www.scopus.com/inward/record.url?scp=84983626328&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84983626328&partnerID=8YFLogxK

U2 - 10.1109/HICSS.1991.184007

DO - 10.1109/HICSS.1991.184007

M3 - Conference article

AN - SCOPUS:84983626328

VL - 2

SP - 448

EP - 457

JO - Proceedings of the Annual Hawaii International Conference on System Sciences

JF - Proceedings of the Annual Hawaii International Conference on System Sciences

SN - 1530-1605

M1 - 184007

ER -