Asynchronous preconditioners and linear solvers
Linear Algebra seminar by Erik Boman, Sandia National Laboratories
Hosted by Institute for Computational and Experimental Research in Mathematics (ICERM), Brown University
Tuesday 10:30 New York (GMT-4)
Recording available
Abstract
Erik Boman discusses preconditioning for asynchronous linear solvers. Inner products create synchronization requirements in Krylov methods, while preconditioners can also improve iterations such as Richardson's method. The talk focuses on asynchronous incomplete factorizations and introduces ATS-ILU, an iterative incomplete LU method with synchronous and asynchronous versions that performs competitively with ParILU.
Topics
Related seminars
Acceleration and Adaptive Selection in Asynchronous Iterative Solvers
Related research
Asynchronous Iterative Methods: From Numerical Solvers to Reinforcement Learning
More on asynchronous iterations
Asynchronous Methods on AMD GPU-Based Systems
Related research