Abstract
In the ongoing efforts targeting the vectorization of linear algebra primitives, sparse matrix-matrix multiplication (SpGEMM) has received considerably less attention than sparse Matrix-Vector multiplication (SpMV). While both are equally important, this disparity can be attributed mainly to the additional formidable challenges raised by SpGEMM.
In this paper, we present a dynamic approach for addressing SpGEMM on the GPU. Our approach works directly on the standard compressed sparse rows (CSR) data format. In comparison to previous SpGEMM implementations, our approach guarantees a homogeneous, load-balanced access pattern to the first input matrix and improves memory access to the second input matrix. It adaptively re-purposes GPU threads during execution and maximizes the time efficient on-chip scratchpad memory can be used. Adhering to a completely deterministic scheduling pattern …
In this paper, we present a dynamic approach for addressing SpGEMM on the GPU. Our approach works directly on the standard compressed sparse rows (CSR) data format. In comparison to previous SpGEMM implementations, our approach guarantees a homogeneous, load-balanced access pattern to the first input matrix and improves memory access to the second input matrix. It adaptively re-purposes GPU threads during execution and maximizes the time efficient on-chip scratchpad memory can be used. Adhering to a completely deterministic scheduling pattern …
Original language | English |
---|---|
Title of host publication | PPoPP '19, Proceedings of the 24th Symposium on Principles and Practice of Parallel Programming |
Place of Publication | New York, NY |
Publisher | Association of Computing Machinery |
Pages | 68-81 |
Number of pages | 14 |
ISBN (Print) | 978-1-4503-6225-2 |
DOIs | |
Publication status | Published - 2019 |
Event | 24th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming - Washington, DC, United States Duration: 16 Feb 2019 → 20 Feb 2019 |
Conference
Conference | 24th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming |
---|---|
Abbreviated title | PPoPP '19 |
Country/Territory | United States |
City | Washington, DC |
Period | 16/02/19 → 20/02/19 |