Improving GPU Matrix Multiplication by Leveraging Bit Level Granularity and Compression
Research output: Contribution to book/Conference proceedings/Anthology/Report › Conference contribution › Contributed › peer-review
Contributors
Abstract
In this paper, we introduce BEAM as a novel approach to perform GPU based matrix multiplication on compressed elements. BEAM allows flexible handling of bit sizes for both input and output elements. First evaluations show promising speedups compared to an uncompressed state-of-the-art matrix multiplication algorithm provided by Nvidia.
Details
| Original language | English |
|---|---|
| Title of host publication | Datenbanksysteme fur Business, Technologie und Web, BTW 2023 |
| Editors | Birgitta Konig-Ries, Stefanie Scherzinger, Wolfgang Lehner, Gottfried Vossen |
| Publisher | Gesellschaft fur Informatik (GI) |
| Pages | 763-772 |
| Number of pages | 10 |
| ISBN (electronic) | 9783885797258 |
| Publication status | Published - 2023 |
| Peer-reviewed | Yes |
Publication series
| Series | Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI) |
|---|---|
| Volume | P-331 |
| ISSN | 1617-5468 |
Symposium
| Title | 20th Conference on Database Systems for Business, Technology and Web |
|---|---|
| Abbreviated title | BTW 2023 |
| Conference number | 20 |
| Duration | 6 - 10 March 2023 |
| Location | Technische Universität Dresden |
| City | Dresden |
| Country | Germany |
External IDs
| ORCID | /0000-0001-8107-2775/work/194824066 |
|---|
Keywords
ASJC Scopus subject areas
Keywords
- GPU, Matrix multiplication