Lecture 20: Cannon’s Algorithm

../_images/L20-title.png Lecture 20 slides Lecture 20 panopto Podcast

(Originally recorded 2019-06-06)

We conclude the course by coming full circle and looking at the (basic) algorithm underlying the benchmarks used by the Top500 list to rank the world’s most powerful supercomputers: matrix-matrix product. In particular, we present Cannon’s algorithm and its realization using MPI.