### Abstract

Scheduling dependent tasks is one of the most challenging versions of the scheduling problem in parallel and distributed systems. It is known to be computationally intractable in its general form as well as several restricted cases. As a result, researchers have studied restricted forms of the problem by constraining either the task graph representing the parallel tasks or the computer model. Also, in an attempt to solve the problem in the general case, a number of heuristics have been developed. In this paper, we study the scheduling problem for a fixed number of processors m. In the proposed work, we approach the problem by recursively reducing the m-processor scheduling to (m-1)-processor scheduling until we apply the optimal two-processor scheduling algorithm when m equals two. This is accomplished by identifying a maximal chain C in the task graph G and merging the (m-1) processor scheduling of (G-C) and the 1-processor scheduling of C. A number of experiments were conducted to compare the suggested approach with the standard list-scheduling algorithm. Based on the outcome of the conducted experiments, the proposed algorithms outperformed or matched the performance of the list heuristic almost all the time.

Original language | English (US) |
---|---|

Title of host publication | Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems |

Editors | T. Gonzalez |

Pages | 429-436 |

Number of pages | 8 |

Volume | 16 |

State | Published - 2004 |

Event | Proceedings of the 16th IASTED International Conference on Parallel and Distributed Computing and Systems - Cambridge, MA, United States Duration: Nov 9 2004 → Nov 11 2004 |

### Other

Other | Proceedings of the 16th IASTED International Conference on Parallel and Distributed Computing and Systems |
---|---|

Country | United States |

City | Cambridge, MA |

Period | 11/9/04 → 11/11/04 |

### Fingerprint

### Keywords

- Heuristics
- Optimal algorithms
- Parallel Processing
- Task Scheduling

### ASJC Scopus subject areas

- Engineering(all)

### Cite this

*Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems*(Vol. 16, pp. 429-436). [439-063]

**A maximal chain approach for scheduling tasks in a multiprocessor system.** / Pawaskar, Sachin; Ali, Hesham H.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems.*vol. 16, 439-063, pp. 429-436, Proceedings of the 16th IASTED International Conference on Parallel and Distributed Computing and Systems, Cambridge, MA, United States, 11/9/04.

}

TY - GEN

T1 - A maximal chain approach for scheduling tasks in a multiprocessor system

AU - Pawaskar, Sachin

AU - Ali, Hesham H

PY - 2004

Y1 - 2004

N2 - Scheduling dependent tasks is one of the most challenging versions of the scheduling problem in parallel and distributed systems. It is known to be computationally intractable in its general form as well as several restricted cases. As a result, researchers have studied restricted forms of the problem by constraining either the task graph representing the parallel tasks or the computer model. Also, in an attempt to solve the problem in the general case, a number of heuristics have been developed. In this paper, we study the scheduling problem for a fixed number of processors m. In the proposed work, we approach the problem by recursively reducing the m-processor scheduling to (m-1)-processor scheduling until we apply the optimal two-processor scheduling algorithm when m equals two. This is accomplished by identifying a maximal chain C in the task graph G and merging the (m-1) processor scheduling of (G-C) and the 1-processor scheduling of C. A number of experiments were conducted to compare the suggested approach with the standard list-scheduling algorithm. Based on the outcome of the conducted experiments, the proposed algorithms outperformed or matched the performance of the list heuristic almost all the time.

AB - Scheduling dependent tasks is one of the most challenging versions of the scheduling problem in parallel and distributed systems. It is known to be computationally intractable in its general form as well as several restricted cases. As a result, researchers have studied restricted forms of the problem by constraining either the task graph representing the parallel tasks or the computer model. Also, in an attempt to solve the problem in the general case, a number of heuristics have been developed. In this paper, we study the scheduling problem for a fixed number of processors m. In the proposed work, we approach the problem by recursively reducing the m-processor scheduling to (m-1)-processor scheduling until we apply the optimal two-processor scheduling algorithm when m equals two. This is accomplished by identifying a maximal chain C in the task graph G and merging the (m-1) processor scheduling of (G-C) and the 1-processor scheduling of C. A number of experiments were conducted to compare the suggested approach with the standard list-scheduling algorithm. Based on the outcome of the conducted experiments, the proposed algorithms outperformed or matched the performance of the list heuristic almost all the time.

KW - Heuristics

KW - Optimal algorithms

KW - Parallel Processing

KW - Task Scheduling

UR - http://www.scopus.com/inward/record.url?scp=11844302831&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=11844302831&partnerID=8YFLogxK

M3 - Conference contribution

VL - 16

SP - 429

EP - 436

BT - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems

A2 - Gonzalez, T.

ER -