Exaflop/s: The why and the how
Open Access
- 31 March 2011
- journal article
- review article
- Published by Cellule MathDoc/Centre Mersenne in Comptes Rendus Mecanique
- Vol. 339 (2-3), 70-77
- https://doi.org/10.1016/j.crme.2010.11.002
Abstract
The best paths to the exascale summit are debatable, but all are narrow and treacherous, constrained by fundamental laws of physics, capital cost, operating cost, power requirements, programmability, and reliability. Many scientific and engineering applications force the modeling community to attempt to scale this summit. Drawing on vendor projections and experiences with scientific codes on contemporary platforms, we outline the challenges and propose roles and essential adaptations for mathematical modelers in one of the great global scientific quests the next decade. © 2010Keywords
This publication has 8 references indexed in Scilit:
- Fast iterative solution of large sparse linear systems on geographically separated clustersThe International Journal of High Performance Computing Applications, 2011
- Partial Differential Equation-Based Applications and Solvers At Extreme ScaleThe International Journal of High Performance Computing Applications, 2009
- A view of the parallel computing landscapeCommunications of the ACM, 2009
- Communication-optimal iterative methodsJournal of Physics: Conference Series, 2009
- Asynchronous Fast Adaptive Composite-Grid Methods for Elliptic Problems: Theoretical FoundationsSIAM Journal on Numerical Analysis, 2004
- High-performance parallel implicit CFDParallel Computing, 2001
- A bridging model for parallel computationCommunications of the ACM, 1990
- Chaotic relaxationLinear Algebra and its Applications, 1969