Threads POSIX

Les threads POSIX, souvent appelés pthreads, sont un sous-standard de la norme POSIX décrivant une interface de programmation permettant de gérer des threads. Il s'agit du standard IEEE Std 1003.1c-1995 (POSIX.1c, Threads extensions). Cette interface est disponible sur la plupart des systèmes Unix modernes, par exemple Linux, les différentes variantes modernes de BSD, Mac OS X et Solaris. Elle n'est pas disponible nativement sous Microsoft Windows mais il existe plusieurs implémentations dont une de Microsoft. Spécification Le standard définit un ensemble de types de données, de fonctions et de constantes en langage C. Le fichier à inclure dans les sources en C se nomme pthread.h. Une centaine de fonctions sont ainsi définies. La bibliothèque dispose notamment de fonctions permettant la création et la destruction de threads, l'ordonnancement des tâches, leur synchronisation, la gestion des signaux et des données partagées. Notes et références Bibliog

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées (3)

Publications associées

Chargement

Personnes associées

Chargement

Unités associées

Chargement

Concepts associés (13)

Concepts associés

Chargement

Cours associés (8)

Cours associés

Chargement

Séances de cours associées (11)

Séances de cours associées

Chargement

Deadlock Immunity: Enabling Systems To Defend Against Deadlocks

George Candea, Horatiu Jula, Daniel Tralamazza, Cristian Zamfir

Deadlock immunity is a property by which programs, once afflicted by a given deadlock, develop resistance against future occurrences of that and similar deadlocks. We describe a technique that enables programs to automatically gain such immunity without assistance from programmers or users. We implemented the technique for both Java and POSIX threads and evaluated it with several real systems, including MySQL, JBoss, SQLite, Apache ActiveMQ, Limewire, and Java JDK. The results demonstrate effectiveness against real deadlock bugs, while incurring modest performance overhead and scaling to 1024 threads. We therefore conclude that deadlock immunity offers programmers and users an attractive tool for coping with elusive deadlocks.

2008

Chargement

OpenMP for Networks of SMPs

Willy Zwaenepoel

In this paper, we present the first system that implement OpenMP on network of shared-memory multiprocessors. This system enables the program to rely on a single, standard, shared-memory API for parallelization within a multiprocessor and between multiprocessors. It is implemented via a translator that convert OpenMP directives to appropriate calls to a modified version of the TreadMarks software distributed shared memory(SDSM) system. In contrast to previous SDSM systems for SMPs, the modified TreadMarks system uses POSIX threads for parallelism within an SMP mode. This approach greatly simplifies the changes required to the SDSM in order to exploit the intra-mode hardware shared memory. We present performance results for seven applications (Barnes-Hut, CLU and Water from SPLASH-2, 3D-FFT from NAS, Red-Black SOR, TSP and MGS) running on an SP2 with four four-processor SMP nodes. A comparison between the thread implementation and the original implementation of TreadMarks shows that using the hardware shared memory within an SMP node significantly achieves speedups that are up to 30% better than the original versions. We also compare SDSM against message passing. Overall, the speedups of multithreaded RreadMarks programs are within 7-30% of the MPI versions.

2000

Chargement

OpenMP on Networks of SMPs

Willy Zwaenepoel

In this paper, we present the first system that implements OpenMP on a network of shared-memory multiprocessors. This system enables the programmer to rely on a single, standard, shared-memory API for parallelization within a multiprocessor and between multiprocessors. It is implemented via a translator that converts OpenMP directives to appropriate calls to a modified version of the TreadMarks software distributed memory system (SDSM). In contrast to previous SDSM systems for SMPs, the modified TreadMarks uses POSIX threads for parallelism within an SMP node. This approach greatly simplifies the changes required to the SDSM in order to exploit the intra-node hardware shared memory. We present performance results for six applications (SPLASH-2 Barnes-Hut andWater, NAS 3D-FFT, SOR, TSP and MGS) running on an SP2 with four four-processor SMP nodes. A comparison between the threaded implementation and the original implementation of TreadMarks shows that using the hardware shared memory within an SMP node significantly reduces the amount of data and the number of messages transmitted between nodes, and consequently achieves speedups up to 30% better than the original versions. We also compare SDSM against message passing. Overall, the speedups of multithreaded TreadMarks programs are within 7–30% of the MPI versions.

1999

Chargement

CS-307: Introduction to multiprocessor architecture

Multiprocessors are a core component in all types of computing infrastructure, from phones to datacenters. This course will build on the prerequisites of processor design and concurrency to introduce the essential technologies required to combine multiple processing elements into a single computer.

CS-472: Design technologies for integrated systems

Hardware compilation is the process of transforming specialized hardware description languages into circuit descriptions, which are iteratively refined, detailed and optimized. The course presents algorithms, tools and methods for hardware compilation and logic synthesis.

CS-323: Introduction to operating systems

Introduction to basic concepts of operating systems.

Thread (informatique)

thumb|Un processus avec deux threads. Un thread ou fil (traduction normalisés par ISO/CEI 2382-7:2000 (autres appellations connues : processus léger, fil d'exécution, fil d'instruction, processus all

Bibliothèque standard du C

La bibliothèque standard du C est une collection maintenant normalisée d'en-têtes et de routines utilisées pour implémenter des opérations courantes, telles que les entrées/sorties et la gestion des

OpenMP

OpenMP (Open Multi-Processing) est une interface de programmation pour le calcul parallèle sur architecture à mémoire partagée. Cette API est prise en charge par de nombreuses plateformes, incluant GN