Hirondml: Fair Threads Migrations for Objective Caml
| Autor: | Vivien Ravet, Emmanuel Chailloux, Julien Verlaguet |
|---|---|
| Přispěvatelé: | Algorithmes, Programmes et Résolution (APR), Laboratoire d'Informatique de Paris 6 (LIP6), Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS) |
| Jazyk: | angličtina |
| Rok vydání: | 2008 |
| Předmět: |
Computer science
Readers–writers problem Distributed computing functional programming 02 engineering and technology Thread (computing) 01 natural sciences Gang scheduling [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] Theoretical Computer Science Scheduling (computing) Global variable Green threads 0103 physical sciences 0202 electrical engineering electronic engineering information engineering Objective Caml distributed programming ComputingMilieux_MISCELLANEOUS 010302 applied physics Copying [INFO.INFO-PL]Computer Science [cs]/Programming Languages [cs.PL] threads migration Local variable 020202 computer hardware & architecture Fair Threads Hardware and Architecture [INFO.INFO-DC]Computer Science [cs]/Distributed Parallel and Cluster Computing [cs.DC] Software |
| Zdroj: | Parallel Processing Letters Parallel Processing Letters, 2008, 18 (1), pp.55-69. ⟨10.1142/S0129626408003223⟩ Parallel Processing Letters, World Scientific Publishing, 2008, 18 (1), pp.55-69. ⟨10.1142/S0129626408003223⟩ |
| ISSN: | 0129-6264 1793-642X |
| Popis: | In this paper, we present HirondML, an Objective Caml library implementing migrating threads. Our library is based on a simplified implementation of the Fair Threads, a reactive threads system originally developed at INRIA. The Fair Threads scheduling policy is based on mixing cooperative and preemptive threads. We designed a thread migration system based on the cooperativeness of Fair Threads. To minimize data copying, we adopted an original rebinding policy distinguishing local variables, which are copied, from global variables, which are rebound when a migration takes place. Such a choice allows two styles of programming, by copy or by sharing. This will be illustrated by different distributed applications. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|