Compact DAG Representation and its Symbolic Scheduling
Identifieur interne : 006C23 ( Main/Merge ); précédent : 006C22; suivant : 006C24Compact DAG Representation and its Symbolic Scheduling
Auteurs : Michel Cosnard ; Emmanuel Jeannot ; Tao YangSource :
- Journal of Parallel and Distributed Computing (JPDC) ; 2004.
English descriptors
Abstract
Task graph scheduling has been found effective in performance prediction and optimization of parallel applications. A number of static scheduling algorithms have been proposed for task graph execution on parallel machines. Such an approach cannot be adapted to changes in values of program parameters and the number of processors and it also cannot handle large task graphs. In this paper, we model parallel computation using parameterized task graphs which represent coarse-grain parallelism independent of the problem size. We present a symbolic scheduling algorithm for a parameterized task graph which first derives linear clusters and then assigns task clusters to processors. The runtime system executes clusters on each processor in a multi-threaded fashion. The experiments using various scientific computing kernel benchmarks show that our method delivers compact and symbolic schedules with performance highly competitive to static approaches.
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: 003C77
- to stream Crin, to step Curation: 003C77
- to stream Crin, to step Checkpoint: 000748
Links to Exploration step
CRIN:cosnard03aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="336">Compact DAG Representation and its Symbolic Scheduling</title></titleStmt><publicationStmt><idno type="RBID">CRIN:cosnard03a</idno><date when="2004" year="2004">2004</date><idno type="wicri:Area/Crin/Corpus">003C77</idno><idno type="wicri:Area/Crin/Curation">003C77</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">003C77</idno><idno type="wicri:Area/Crin/Checkpoint">000748</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">000748</idno><idno type="wicri:Area/Main/Merge">006C23</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Compact DAG Representation and its Symbolic Scheduling</title><author><name sortKey="Cosnard, Michel" sort="Cosnard, Michel" uniqKey="Cosnard M" first="Michel" last="Cosnard">Michel Cosnard</name></author><author><name sortKey="Jeannot, Emmanuel" sort="Jeannot, Emmanuel" uniqKey="Jeannot E" first="Emmanuel" last="Jeannot">Emmanuel Jeannot</name></author><author><name sortKey="Yang, Tao" sort="Yang, Tao" uniqKey="Yang T" first="Tao" last="Yang">Tao Yang</name></author></analytic><series><title level="j">Journal of Parallel and Distributed Computing (JPDC)</title><imprint><date when="2004" type="published">2004</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>automatic parallelization</term><term>linear clustering</term><term>parameterized task graph</term><term>static schedduling</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en" wicri:score="1995">Task graph scheduling has been found effective in performance prediction and optimization of parallel applications. A number of static scheduling algorithms have been proposed for task graph execution on parallel machines. Such an approach cannot be adapted to changes in values of program parameters and the number of processors and it also cannot handle large task graphs. In this paper, we model parallel computation using parameterized task graphs which represent coarse-grain parallelism independent of the problem size. We present a symbolic scheduling algorithm for a parameterized task graph which first derives linear clusters and then assigns task clusters to processors. The runtime system executes clusters on each processor in a multi-threaded fashion. The experiments using various scientific computing kernel benchmarks show that our method delivers compact and symbolic schedules with performance highly competitive to static approaches.</div></front></TEI></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 006C23 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 006C23 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= Main
|étape= Merge
|type= RBID
|clé= CRIN:cosnard03a
|texte= Compact DAG Representation and its Symbolic Scheduling
}}
| This area was generated with Dilib version V0.6.33. |  |