Improving pipelined time stepping algorithm for distributed memory multicomputers

Time stepping algorithm with spatial parallelisation is commonly used to solve time dependent partial differential equations. Computation in each time step is carried out using all processors available before sequentially advancing to the next time step. In cases where few spatial components are inv...

Full description

Bibliographic Details
Main Authors:	Ng, Kok Fu, Norhashidah Hj. Mohd. Ali
Format:	Article
Language:	English
Published:	Universiti Kebangsaan Malaysia 2010
Online Access:	http://journalarticle.ukm.my/7451/ http://journalarticle.ukm.my/7451/ http://journalarticle.ukm.my/7451/1/26_Ayiesah.pdf

id	ukm-7451
recordtype	eprints
spelling	ukm-74512016-12-14T06:44:08Z http://journalarticle.ukm.my/7451/ Improving pipelined time stepping algorithm for distributed memory multicomputers Ng, Kok Fu Norhashidah Hj. Mohd. Ali, Time stepping algorithm with spatial parallelisation is commonly used to solve time dependent partial differential equations. Computation in each time step is carried out using all processors available before sequentially advancing to the next time step. In cases where few spatial components are involved and there are relatively many processors available for use, this will result in fine granularity and decreased scalability. Naturally one alternative is to parallelise the temporal domain. Several time parallelisation algorithms have been suggested for the past two decades. One of them is the pipelined iterations across time steps. In this pipelined time stepping method, communication however is extensive between time steps during the pipelining process. This causes a decrease in performance on distributed memory environment which often has high message latency. We present a modified pipelined time stepping algorithm based on delayed pipelining and reduced communication strategies to improve overall execution time on a distributed memory environment using MPI. Our goal is to reduce the inter-time step communications while providing adequate information for the next time step to converge. Numerical result confirms that the improved algorithm is faster than the original pipelined algorithm and sequential time stepping algorithm with spatial parallelisation alone. The improved algorithm is most beneficial for fine granularity time dependent problems with limited spatial parallelisation. Universiti Kebangsaan Malaysia 2010-12 Article PeerReviewed application/pdf en http://journalarticle.ukm.my/7451/1/26_Ayiesah.pdf Ng, Kok Fu and Norhashidah Hj. Mohd. Ali, (2010) Improving pipelined time stepping algorithm for distributed memory multicomputers. Sains Malaysiana, 39 (6). pp. 1041-1048. ISSN 0126-6039 http://www.ukm.my/jsm/english_journals/vol39num6_2010/contentsVol39num6_2010.html
repository_type	Digital Repository
institution_category	Local University
institution	Universiti Kebangasaan Malaysia
building	UKM Institutional Repository
collection	Online Access
language	English
description	Time stepping algorithm with spatial parallelisation is commonly used to solve time dependent partial differential equations. Computation in each time step is carried out using all processors available before sequentially advancing to the next time step. In cases where few spatial components are involved and there are relatively many processors available for use, this will result in fine granularity and decreased scalability. Naturally one alternative is to parallelise the temporal domain. Several time parallelisation algorithms have been suggested for the past two decades. One of them is the pipelined iterations across time steps. In this pipelined time stepping method, communication however is extensive between time steps during the pipelining process. This causes a decrease in performance on distributed memory environment which often has high message latency. We present a modified pipelined time stepping algorithm based on delayed pipelining and reduced communication strategies to improve overall execution time on a distributed memory environment using MPI. Our goal is to reduce the inter-time step communications while providing adequate information for the next time step to converge. Numerical result confirms that the improved algorithm is faster than the original pipelined algorithm and sequential time stepping algorithm with spatial parallelisation alone. The improved algorithm is most beneficial for fine granularity time dependent problems with limited spatial parallelisation.
format	Article
author	Ng, Kok Fu Norhashidah Hj. Mohd. Ali,
spellingShingle	Ng, Kok Fu Norhashidah Hj. Mohd. Ali, Improving pipelined time stepping algorithm for distributed memory multicomputers
author_facet	Ng, Kok Fu Norhashidah Hj. Mohd. Ali,
author_sort	Ng, Kok Fu
title	Improving pipelined time stepping algorithm for distributed memory multicomputers
title_short	Improving pipelined time stepping algorithm for distributed memory multicomputers
title_full	Improving pipelined time stepping algorithm for distributed memory multicomputers
title_fullStr	Improving pipelined time stepping algorithm for distributed memory multicomputers
title_full_unstemmed	Improving pipelined time stepping algorithm for distributed memory multicomputers
title_sort	improving pipelined time stepping algorithm for distributed memory multicomputers
publisher	Universiti Kebangsaan Malaysia
publishDate	2010
url	http://journalarticle.ukm.my/7451/ http://journalarticle.ukm.my/7451/ http://journalarticle.ukm.my/7451/1/26_Ayiesah.pdf
first_indexed	2023-09-18T19:49:45Z
last_indexed	2023-09-18T19:49:45Z
_version_	1777406140163817472

Improving pipelined time stepping algorithm for distributed memory multicomputers

Similar Items