Video: ECP Project for Linear Algebra Targeting Exascale

Print Friendly, PDF & Email

Jacob Kurzak

In this video, Jakub Kurzak, Research Assistant Professor at the University of Tennessee’s Innovative Computing Laboratory, discusses the Software for Linear Algebra Targeting Exascale (SLATE) project.

The objective of SLATE is to provide basic dense matrix operations such as matrix multiplication, rank-k update, triangular solve, as well as linear systems solvers, least square solvers, singular value and eigenvalue solvers, in support of the ECP efforts to build a capable exascale computing ecosystem.

Ultimately, SLATE hopes to replace the venerable Scalable Linear Algebra PACKage (ScaLAPACK) library, which has become the industry standard for dense linear algebra operations in distributed memory environments, but unfortunately is a library at the end of its lifecycle due to the growing need to support hardware accelerators, which are an integral part of today’s high-performance computing hardware infrastructure.

The Exascale Computing Project (ECPis focused on accelerating the delivery of a capable exascale computing ecosystem that delivers 50 times more computational science and data analytic application power than possible with DOE HPC systems such as Titan (ORNL) and Sequoia (LLNL). With the goal to launch a US exascale ecosystem by 2021, the ECP will have profound effects on the American people and the world.

Sign up for our insideHPC Newsletter