TaskLocalRandom: A Statistically Sound Substitute to Pseudorandom Number Generation in Parallel Java Tasks Frameworks
Uloženo v:
| Název: | TaskLocalRandom: A Statistically Sound Substitute to Pseudorandom Number Generation in Parallel Java Tasks Frameworks |
|---|---|
| Autoři: | Passerat-Palmbach, Jonathan, Mazel, Claude, Hill, David R.C. |
| Přispěvatelé: | Institut Supérieur d'Informatique, de Modélisation et de leurs Applications (ISIMA), Université Blaise Pascal - Clermont-Ferrand 2 (UBP)-Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), Clermont Université, Laboratoire d'Informatique, de Modélisation et d'optimisation des Systèmes (LIMOS), Université Blaise Pascal - Clermont-Ferrand 2 (UBP)-Université d'Auvergne - Clermont-Ferrand I (UdA)-SIGMA Clermont (SIGMA Clermont)-Ecole Nationale Supérieure des Mines de St Etienne (ENSM ST-ETIENNE)-Centre National de la Recherche Scientifique (CNRS), Université Blaise Pascal - Clermont-Ferrand 2 (UBP), CPER Auvergne 2007-2014 Innov@Pôle - Auvergrid, FEDER |
| Zdroj: | ISSN: 1532-0626 ; EISSN: 1532-0634. |
| Informace o vydavateli: | CCSD Wiley |
| Rok vydání: | 2014 |
| Sbírka: | HAL Clermont Auvergne (Université Blaise Pascal Clermont-Ferrand / Université d'Auvergne) |
| Témata: | Java, Threads, Tasks, Pseudorandom Number Generation, Parallelization of Simu- lation, Software Development Tools and Support, Object Oriented Programming & Design, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation |
| Popis: | International audience ; Several software efforts have been produced over the past few years in variousprogramming languages to help developers handle pseudorandom streams partitioning. Paralleland Distributed Stochastic Simulations (PDSS) can obviously benefit from this kind of high-leveltools. The latest release of the Java Development Kit (JDK 7) tries to tackle this problem byproviding facilities to partition a pseudorandom stream across various threads thanks to the newclass ThreadLocalRandom. Meanwhile, Java 7 offers a framework to split a problem in a divide andconquer way through the new class called ForkJoinPool. As any other Java Thread Pool, ForkJoinexploits threads as workers and manipulates the tasks that will be run on the workers. In Thread-LocalRandom, pseudorandom number generation is handled at a thread level. As a consequence,a scientific application taking advantage of a Java Thread Pool to parallelize its computation maysuffer from a bad pseudorandom stream partitioning due to the behaviour of ThreadLocalRandom.The present work introduces TaskLocalRandom, a task-level alternative to ThreadLocalRandomthat solves this partitioning problem and assigns an independent pseudorandom stream to eachtask run in the thread pool. TaskLocalRandom is compatible with existing Java thread pools suchas Executors or ForkJoin. |
| Druh dokumentu: | article in journal/newspaper |
| Jazyk: | English |
| DOI: | 10.1002/cpe.3214 |
| Dostupnost: | https://inria.hal.science/hal-01098598 https://inria.hal.science/hal-01098598v1/document https://inria.hal.science/hal-01098598v1/file/ccpe2013_passerat-palmbach_etal_frree.pdf https://doi.org/10.1002/cpe.3214 |
| Rights: | http://creativecommons.org/licenses/by/ ; info:eu-repo/semantics/OpenAccess |
| Přístupové číslo: | edsbas.3BAE22E4 |
| Databáze: | BASE |
Nájsť tento článok vo Web of Science Buďte první, kdo okomentuje tento záznam!