Mixed task and data parallel executions in general linear methods
Thomas Rauber,Gudula Rünger +1 more
TL;DR: This paper studies mixed task and data parallel implementations for general linear methods realized using a library for multiprocessor task programming and shows good efficiency results.
read more
Abstract: On many parallel target platforms it can be advantageous to implement parallel applications as a collection of multiprocessor tasks that are concurrently executed and are internally implemented with fine-grain SPMD parallelism. A class of applications which can benefit from this programming style are methods for solving systems of ordinary differential equations. Many recent solvers have been designed with an additional potential of method parallelism, but the actual effectiveness of mixed task and data parallelism depends on the specific communication and computation requirements imposed by the equation to be solved. In this paper we study mixed task and data parallel implementations for general linear methods realized using a library for multiprocessor task programming. Experiments on a number of different platforms show good efficiency results.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Work-stealing for mixed-mode parallelism by deterministic team-building
Martin Wimmer,Jesper Larsson Träff +1 more
- 04 Jun 2011
TL;DR: Work-stealing with deterministic team-building as mentioned in this paper is an extension of the classical workstealing algorithm to handle tightly coupled data parallel tasks that can require any number of threads r ≥ 1 for their execution.
17
•Posted Content
Work-stealing for mixed-mode parallelism by deterministic team-building
TL;DR: This work shows how to extend classical work-stealing to deal with tightly coupled data parallel tasks that can require any number of threads r ≥ 1 for their execution, and term this extension work-Stealing with deterministic team-building.
14
An Extended Work-Stealing Framework for Mixed-Mode Parallel Applications
Martin Wimmer,Jesper Larsson Träff +1 more
- 16 May 2011
TL;DR: This paper presents a shared-memory programming framework that allows tasks to dynamically spawn subtasks with a given degree of parallelism for implementing tightly coupled parallel parts of the algorithm, and presents a new algorithm for work-stealing with deterministic team-building.
5
References
•Book
Solving Ordinary Differential Equations I: Nonstiff Problems
Ernst Hairer,Syvert P. Nørsett,Gerhard Wanner +2 more
- 01 Jan 1987
TL;DR: In this paper, the authors describe the historical development of the classical theory of linear methods for solving nonstiff ODEs and present a modern treatment of Runge-Kutta and extrapolation methods.
4.2K
The numerical analysis of ordinary differential equations: Runge-Kutta and general linear methods
TL;DR: The Euler Method and its Generalizations Analysis of Runge-Kutta Methods General Linear Methods Bibliography.
1.4K
•Book
Solving ordinary differential equations I (2nd revised. ed.): nonstiff problems
E. Hairer,Syvert P. Nørsett,G. Wanner +2 more
- 01 Jul 1993
1.3K
•Book
The numerical analysis of ordinary differential equations
John C. Butcher
- 01 Jan 1987
TL;DR: In this article, a temperature control system for individual rooms within a home is disclosed having a pair of thermostatically controlled switches settable to "occupied" and "unoccupied" temperatures respectively with a clock controlled switching arrangement for determining which switch is effective during predetermined hours of the day.
800