Multigrid at Scale?
Thursday, February 18, 2016
3:30PM – 5PM
Multigrid and multilevel iterative algorithms are often the method of choice for the solution of large-scale systems of linear equations arising from discretisation of partial differential equations using finite element or finite difference methods. The method involves a number of components including smoothing relaxation, coarse grid solve, prolongation and restriction operators between grids in the multilevel hierarchy, and the convergence behavior of the method has been extensively analysed in the context of standard computer architectures. However, comparatively little is known about the resilience or fault-tolerance of the algorithm on next generation hardware architectures which are expected to suffer from frequent data corruption and hardware failures. We will address this issue and present some of the results of our recent work showing that the issue is anything but clear.
This is joint work with Christian Glusa (Brown University).
Mark Ainsworth obtained his PhD in Mathematics at Durham University in the United Kingdom. Prior to moving to Brown, he held the 1825 Chair in Mathematics at Strathclyde University and was Director of
NAIS , a joint centre between the Universities of Edinburgh, Heriot-Watt and Strathclyde, and Edinburgh Parallel Computing Centre to develop UK capacity in high performance computing and numerical analysis.
Learn More About Mark Ainsworth
Hosted by Leszek Demkowicz