Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization

Heejo Lee, Jong Kim, Sung Je Hong, Sunggu Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Block-oriented sparse Cholesky factorization decomposes a sparse matrix into rectangular sub-blocks; each block can then be handled as a computational unit in order to increase data reuse in a hierarchical memory system. Also, the factorization method increases the degree of concurrency with the reduction of communication volumes so that it performs more efficiently on a distributed-memory multiprocessor system than the customary column-oriented factorization method. But until now, mapping of blocks to processors has been designed for load balance with restricted communication patterns. In this paper, we represent tasks using a block dependency DAG that shows the execution behavior of block sparse Cholesky factorization in a distributed-memory system. Since the characteristics of tasks for the block Cholesky factorization are different from those of the conventional parallel task model, we propose a new task scheduling algorithm using a block dependency DAG. The proposed algorithm consists of two stages: early-start clustering, and affined cluster mapping. The early-start clustering stage is used to cluster tasks with preserving the earliest start time of a task without limiting parallelism. After task clustering, the affined cluster mapping stage allocates clusters to processors considering both communication cost and load balance. Experimental results on the Fujitsu parallel system show that the proposed task scheduling approach outperforms other processor mapping methods.

Original languageEnglish
Title of host publicationProceedings of the ACM Symposium on Applied Computing
Pages641-648
Number of pages8
Volume2
DOIs
Publication statusPublished - 2000 Dec 1
Externally publishedYes
Event2000 ACM Symposium on Applied Computing, SAC 2000 - Como, Italy
Duration: 2000 Mar 192000 Mar 21

Other

Other2000 ACM Symposium on Applied Computing, SAC 2000
CountryItaly
CityComo
Period00/3/1900/3/21

Fingerprint

Factorization
Scheduling
Data storage equipment
Communication
Scheduling algorithms
Costs

Keywords

  • Block-oriented Cholesky factorization
  • Directed acyclic graph
  • Parallel sparse matrix factorization
  • Task scheduling

ASJC Scopus subject areas

  • Software

Cite this

Lee, H., Kim, J., Hong, S. J., & Lee, S. (2000). Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization. In Proceedings of the ACM Symposium on Applied Computing (Vol. 2, pp. 641-648) https://doi.org/10.1145/338407.338535

Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization. / Lee, Heejo; Kim, Jong; Hong, Sung Je; Lee, Sunggu.

Proceedings of the ACM Symposium on Applied Computing. Vol. 2 2000. p. 641-648.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lee, H, Kim, J, Hong, SJ & Lee, S 2000, Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization. in Proceedings of the ACM Symposium on Applied Computing. vol. 2, pp. 641-648, 2000 ACM Symposium on Applied Computing, SAC 2000, Como, Italy, 00/3/19. https://doi.org/10.1145/338407.338535
Lee H, Kim J, Hong SJ, Lee S. Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization. In Proceedings of the ACM Symposium on Applied Computing. Vol. 2. 2000. p. 641-648 https://doi.org/10.1145/338407.338535
Lee, Heejo ; Kim, Jong ; Hong, Sung Je ; Lee, Sunggu. / Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization. Proceedings of the ACM Symposium on Applied Computing. Vol. 2 2000. pp. 641-648
@inproceedings{aa43081ef5fe4f2cab4af4ce59c3f20e,
title = "Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization",
abstract = "Block-oriented sparse Cholesky factorization decomposes a sparse matrix into rectangular sub-blocks; each block can then be handled as a computational unit in order to increase data reuse in a hierarchical memory system. Also, the factorization method increases the degree of concurrency with the reduction of communication volumes so that it performs more efficiently on a distributed-memory multiprocessor system than the customary column-oriented factorization method. But until now, mapping of blocks to processors has been designed for load balance with restricted communication patterns. In this paper, we represent tasks using a block dependency DAG that shows the execution behavior of block sparse Cholesky factorization in a distributed-memory system. Since the characteristics of tasks for the block Cholesky factorization are different from those of the conventional parallel task model, we propose a new task scheduling algorithm using a block dependency DAG. The proposed algorithm consists of two stages: early-start clustering, and affined cluster mapping. The early-start clustering stage is used to cluster tasks with preserving the earliest start time of a task without limiting parallelism. After task clustering, the affined cluster mapping stage allocates clusters to processors considering both communication cost and load balance. Experimental results on the Fujitsu parallel system show that the proposed task scheduling approach outperforms other processor mapping methods.",
keywords = "Block-oriented Cholesky factorization, Directed acyclic graph, Parallel sparse matrix factorization, Task scheduling",
author = "Heejo Lee and Jong Kim and Hong, {Sung Je} and Sunggu Lee",
year = "2000",
month = "12",
day = "1",
doi = "10.1145/338407.338535",
language = "English",
isbn = "1581132409",
volume = "2",
pages = "641--648",
booktitle = "Proceedings of the ACM Symposium on Applied Computing",

}

TY - GEN

T1 - Task scheduling using a block dependency DAG for block-oriented sparse Cholesky factorization

AU - Lee, Heejo

AU - Kim, Jong

AU - Hong, Sung Je

AU - Lee, Sunggu

PY - 2000/12/1

Y1 - 2000/12/1

N2 - Block-oriented sparse Cholesky factorization decomposes a sparse matrix into rectangular sub-blocks; each block can then be handled as a computational unit in order to increase data reuse in a hierarchical memory system. Also, the factorization method increases the degree of concurrency with the reduction of communication volumes so that it performs more efficiently on a distributed-memory multiprocessor system than the customary column-oriented factorization method. But until now, mapping of blocks to processors has been designed for load balance with restricted communication patterns. In this paper, we represent tasks using a block dependency DAG that shows the execution behavior of block sparse Cholesky factorization in a distributed-memory system. Since the characteristics of tasks for the block Cholesky factorization are different from those of the conventional parallel task model, we propose a new task scheduling algorithm using a block dependency DAG. The proposed algorithm consists of two stages: early-start clustering, and affined cluster mapping. The early-start clustering stage is used to cluster tasks with preserving the earliest start time of a task without limiting parallelism. After task clustering, the affined cluster mapping stage allocates clusters to processors considering both communication cost and load balance. Experimental results on the Fujitsu parallel system show that the proposed task scheduling approach outperforms other processor mapping methods.

AB - Block-oriented sparse Cholesky factorization decomposes a sparse matrix into rectangular sub-blocks; each block can then be handled as a computational unit in order to increase data reuse in a hierarchical memory system. Also, the factorization method increases the degree of concurrency with the reduction of communication volumes so that it performs more efficiently on a distributed-memory multiprocessor system than the customary column-oriented factorization method. But until now, mapping of blocks to processors has been designed for load balance with restricted communication patterns. In this paper, we represent tasks using a block dependency DAG that shows the execution behavior of block sparse Cholesky factorization in a distributed-memory system. Since the characteristics of tasks for the block Cholesky factorization are different from those of the conventional parallel task model, we propose a new task scheduling algorithm using a block dependency DAG. The proposed algorithm consists of two stages: early-start clustering, and affined cluster mapping. The early-start clustering stage is used to cluster tasks with preserving the earliest start time of a task without limiting parallelism. After task clustering, the affined cluster mapping stage allocates clusters to processors considering both communication cost and load balance. Experimental results on the Fujitsu parallel system show that the proposed task scheduling approach outperforms other processor mapping methods.

KW - Block-oriented Cholesky factorization

KW - Directed acyclic graph

KW - Parallel sparse matrix factorization

KW - Task scheduling

UR - http://www.scopus.com/inward/record.url?scp=0012650251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0012650251&partnerID=8YFLogxK

U2 - 10.1145/338407.338535

DO - 10.1145/338407.338535

M3 - Conference contribution

AN - SCOPUS:0012650251

SN - 1581132409

SN - 9781581132403

VL - 2

SP - 641

EP - 648

BT - Proceedings of the ACM Symposium on Applied Computing

ER -