Modeling the effect of application-specific program transformations on energy and performance improvements of parallel ODE solvers
Authors:
- Thomas RauberTRThomas Rauber
- Gudula RüngerGRGudula Rünger
- 1 April 2021
- journal article
- research article
- Published by Elsevier in Journal of Computational Science
- Vol. 51
Abstract
No abstract availableKeywords
Funding Information
- Bundesministerium für Forschung, Technologie und Raumfahrt
- Bundesministerium für Forschung, Technologie und Raumfahrt (01IH16012A/B)
This publication has 26 references indexed in Scilit:
- Unveiling the performance‐energy trade‐off in iterative linear system solvers for multithreaded processorsConcurrency and Computation: Practice and Experience, 2014
- Energy measurement, modeling, and prediction for processors with frequency scalingThe Journal of Supercomputing, 2014
- Energy efficiency vs. performance of the numerical solution of PDEs: An application study on a low-power ARM-based clusterJournal of Computational Physics, 2013
- Power challenges may end the multicore eraCommunications of the ACM, 2013
- Power-efficient softwareCommunications of the ACM, 2010
- Energy-efficient dynamic task scheduling algorithms for DVS systemsACM Transactions on Embedded Computing Systems, 2008
- Parallel Adams methodsJournal of Computational and Applied Mathematics, 1999
- Massive parallelism across space in ODEsApplied Numerical Mathematics, 1993
- Parallel methods for initial value problemsApplied Numerical Mathematics, 1993
- Parallel iteration of high-order Runge-Kutta methods with stepsize controlJournal of Computational and Applied Mathematics, 1990