Accession Number:
AD1006016
Title:
Scalable High-order Methods for Multi-Scale Problems: Analysis, Algorithms and Application
Descriptive Note:
Technical Report,30 Sep 2012,29 Nov 2016
Corporate Author:
Brown University Providence United States
Personal Author(s):
Report Date:
2016-02-26
Pagination or Media Count:
12.0
Abstract:
It is anticipated that in future generations of massively parallel computer systems a significant portion of processors may suffer from hardware or software faults rendering large-scale computations useless. In this project, the PI and his students address this problem from the algorithmic side, proposing resilient frameworks that can recover and continue the solution with gappy fields from such faults irrespective of their fault origin. In addition to its robustness and resilience, the new framework generalizes previous multiscale and multifidelity approaches in a unified parallel computational framework.