Sparse matrix computation
Abstract
This chapter introduces the parallel sparse matrix-vector computation pattern. It starts with the basic concepts of sparse matrices, which are matrices in which most of the elements are zeros. It introduces the coordinate list format as a flexible representation that does not store zero matrix elements. It then introduces a kernel based on compressed sparse row data storage for sparse matrices. The ELL format with data padding is then introduced as a regularization technique for improved memory coalescing. Finally, the jagged diagonal storage format based on sorting is introduced to smooth out variation from one row to the next row, allowing further reduction of control divergence and padding overhead ...
Get Programming Massively Parallel Processors, 4th Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
