Clustering algorithm for scheduling parallel programs on NOWs with synchronization requirements at the application level

B. R. Arafeh

doi:10.1109/IPDPS.2002.1015567

Clustering algorithm for scheduling parallel programs on NOWs with synchronization requirements at the application level

B. R. Arafeh^*

^*Corresponding author for this work

Computer Science

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

2 Citations (Scopus)

Abstract

In this work, we are interested in developing an efficient heuristic algorithm for scheduling the tasks of a parallel program based on the class of UNC (Unbounded Number of Clusters) scheduling algorithms for clusters of NOWs. The main objective of the proposed UNC algorithm is to consider synchronous communication with deadlock avoidance strategy for inter-task message-passing. The proposed algorithm generates non-linear clusters by traversing the task graph (DAG) once, using the Edge-Zeroing (EZ) technique. The objectives of the clustering algorithm is reducing the parallel time of the program, reducing the communication cost, improving the program computation to communication ratio (PCCR), and avoiding deadlock situations. The algorithm achieves its objectives with a time complexity O(|V|(log|V| + (|E|)²)) using nonlinear clustering in order to avoid more than one pass through the task DAG and to be able to deal with task DAGs with fine, medium, and coarse granularity.

Original language	English
Title of host publication	Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2002
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	625-632
Number of pages	8
ISBN (Print)	0769515738, 9780769515731
DOIs	https://doi.org/10.1109/IPDPS.2002.1015567
Publication status	Published - 2002
Event	16th International Parallel and Distributed Processing Symposium, IPDPS 2002 - Ft. Lauderdale, United States Duration: Apr 15 2002 → Apr 19 2002

Other

Other	16th International Parallel and Distributed Processing Symposium, IPDPS 2002
Country/Territory	United States
City	Ft. Lauderdale
Period	4/15/02 → 4/19/02

ASJC Scopus subject areas

Computer Networks and Communications
Modelling and Simulation

Access to Document

10.1109/IPDPS.2002.1015567

Cite this

Clustering algorithm for scheduling parallel programs on NOWs with synchronization requirements at the application level. / Arafeh, B. R.

Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2002. Institute of Electrical and Electronics Engineers Inc., 2002. p. 625-632 1015567.

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

Arafeh, BR 2002, Clustering algorithm for scheduling parallel programs on NOWs with synchronization requirements at the application level. in Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2002., 1015567, Institute of Electrical and Electronics Engineers Inc., pp. 625-632, 16th International Parallel and Distributed Processing Symposium, IPDPS 2002, Ft. Lauderdale, United States, 4/15/02. https://doi.org/10.1109/IPDPS.2002.1015567

@inproceedings{90e2586859bc485da9a2eb61a4c3c7ed,

title = "Clustering algorithm for scheduling parallel programs on NOWs with synchronization requirements at the application level",

abstract = "In this work, we are interested in developing an efficient heuristic algorithm for scheduling the tasks of a parallel program based on the class of UNC (Unbounded Number of Clusters) scheduling algorithms for clusters of NOWs. The main objective of the proposed UNC algorithm is to consider synchronous communication with deadlock avoidance strategy for inter-task message-passing. The proposed algorithm generates non-linear clusters by traversing the task graph (DAG) once, using the Edge-Zeroing (EZ) technique. The objectives of the clustering algorithm is reducing the parallel time of the program, reducing the communication cost, improving the program computation to communication ratio (PCCR), and avoiding deadlock situations. The algorithm achieves its objectives with a time complexity O(|V|(log|V| + (|E|)2)) using nonlinear clustering in order to avoid more than one pass through the task DAG and to be able to deal with task DAGs with fine, medium, and coarse granularity.",

author = "Arafeh, {B. R.}",

year = "2002",

doi = "10.1109/IPDPS.2002.1015567",

language = "English",

isbn = "0769515738",

pages = "625--632",

booktitle = "Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2002",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "16th International Parallel and Distributed Processing Symposium, IPDPS 2002 ; Conference date: 15-04-2002 Through 19-04-2002",

}

TY - GEN

T1 - Clustering algorithm for scheduling parallel programs on NOWs with synchronization requirements at the application level

AU - Arafeh, B. R.

PY - 2002

Y1 - 2002

N2 - In this work, we are interested in developing an efficient heuristic algorithm for scheduling the tasks of a parallel program based on the class of UNC (Unbounded Number of Clusters) scheduling algorithms for clusters of NOWs. The main objective of the proposed UNC algorithm is to consider synchronous communication with deadlock avoidance strategy for inter-task message-passing. The proposed algorithm generates non-linear clusters by traversing the task graph (DAG) once, using the Edge-Zeroing (EZ) technique. The objectives of the clustering algorithm is reducing the parallel time of the program, reducing the communication cost, improving the program computation to communication ratio (PCCR), and avoiding deadlock situations. The algorithm achieves its objectives with a time complexity O(|V|(log|V| + (|E|)2)) using nonlinear clustering in order to avoid more than one pass through the task DAG and to be able to deal with task DAGs with fine, medium, and coarse granularity.

AB - In this work, we are interested in developing an efficient heuristic algorithm for scheduling the tasks of a parallel program based on the class of UNC (Unbounded Number of Clusters) scheduling algorithms for clusters of NOWs. The main objective of the proposed UNC algorithm is to consider synchronous communication with deadlock avoidance strategy for inter-task message-passing. The proposed algorithm generates non-linear clusters by traversing the task graph (DAG) once, using the Edge-Zeroing (EZ) technique. The objectives of the clustering algorithm is reducing the parallel time of the program, reducing the communication cost, improving the program computation to communication ratio (PCCR), and avoiding deadlock situations. The algorithm achieves its objectives with a time complexity O(|V|(log|V| + (|E|)2)) using nonlinear clustering in order to avoid more than one pass through the task DAG and to be able to deal with task DAGs with fine, medium, and coarse granularity.

UR - http://www.scopus.com/inward/record.url?scp=84966551184&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84966551184&partnerID=8YFLogxK

U2 - 10.1109/IPDPS.2002.1015567

DO - 10.1109/IPDPS.2002.1015567

M3 - Conference contribution

AN - SCOPUS:84966551184

SN - 0769515738

SN - 9780769515731

SP - 625

EP - 632

BT - Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2002

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 16th International Parallel and Distributed Processing Symposium, IPDPS 2002

Y2 - 15 April 2002 through 19 April 2002

ER -

Clustering algorithm for scheduling parallel programs on NOWs with synchronization requirements at the application level

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this