Projects per year
Abstract / Description of output
Sparse matrices and linear algebra are at the heart of scientific simulations. More than 70 sparse matrix storage formats have been developed over the years, targeting a wide range of hardware architectures and matrix types. Each format is developed to exploit the particular strengths of an architecture, or the specific sparsity patterns of matrices, and the choice of the right format can be crucial in order to achieve optimal performance. The adoption of dynamic sparse matrices that can change the underlying data-structure to match the computation at runtime without introducing prohibitive overheads has the potential of optimizing performance through dynamic format selection.In this paper, we introduce Morpheus, a library that provides an efficient abstraction for dynamic sparse matrices. The adoption of dynamic matrices aims to improve the productivity of developers and end-users who do not need to know and understand the implementation specifics of the different formats available, but still want to take advantage of the optimization opportunity to improve the performance of their applications. We demonstrate that by porting HPCG to use Morpheus, and without further code changes, 1) HPCG can now target heterogeneous environments and 2) the performance of the Sparse Matrix-Vector Multiplication (SpMV) kernel is improved up to 2.5× and 7× on CPUs and GPUs respectively, through runtime selection of the best format on each MPI process.
Original language | English |
---|---|
Pages | 47-57 |
Number of pages | 11 |
DOIs | |
Publication status | Published - 2022 |
Event | The International Conference for High Performance Computing, Networking, Storage, and Analysis: P3HPC Workshop - Dallas, United States Duration: 13 Nov 2022 → 13 Nov 2023 https://p3hpc.org/ |
Conference
Conference | The International Conference for High Performance Computing, Networking, Storage, and Analysis |
---|---|
Abbreviated title | SC22 |
Country/Territory | United States |
City | Dallas |
Period | 13/11/22 → 13/11/23 |
Internet address |
Keywords / Materials (for Non-textual outputs)
- sparse matrix storage formats
- generic programming
- dynamic matrices
- performance portability
- productivity
Fingerprint
Dive into the research topics of 'Exploiting dynamic sparse matrices for performance portable linear algebra operations'. Together they form a unique fingerprint.Projects
- 1 Finished
-
Strategic Partnership in Computational Science for Advanced Simulation and Modelling of Engineering Systems - ASiMoV
1/10/18 → 31/03/24
Project: Research
Research output
- 1 Paper
-
An approach to performance portability through generic programming
Hadjigeorgiou, A., Stylianou, C., Weiland, M., Verschuur, D. J. & Finkenrath, J., 16 Apr 2024.Research output: Contribution to conference › Paper › peer-review
File