Abstracts: Jack Dongarra
PLASMA and Scheduling Dense Linear Algebra on Multicore Chips
PLASMA is a framework for a dense linear algebra library for multicore processors. The main objective of PLASMA is to address performance challenges of implementing compute intensive numerical routines on multicore hardware. PLASMA accomplishes this goal by utilizing novel algorithms, data structures and parallel scheduling methods. Here, main concepts behind the design of PLASMA are presented, followed by a discussion of different mechanisms for parallel task scheduling and different methods for expressing parallelism in software.