Extending ginkgo to manage Reconfigurable Hardware-Based Kernels
Revista Colombiana de Computación · 31.12.2024 · 25 (2) · 43-58
DOI: 10.29375/25392115.5276 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista Colombiana de Computación, online erschienen am 31.12.2024. Verfasst von Alejandro Morales-Peña und 1 weiteren Autor:innen. Die Arbeit nennt 10 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
Although heterogeneous systems based on hardware accelerators are a trending topic in the HPC community, exploring the trade-offs of reconfigurable hardware-based ones in linear algebra libraries for high-performance systems, has not been deeply studied. Therefore, in this research, we aim to take advantage of FPGAs' reconfigurability, adaptability, and capacity to reduce power consumption to generate FPGA-based kernels in Ginkgo, a specialized high-performance linear algebra library for many-core systems. We generated 3 FPGA-based kernels for the CSR, SELLP, and SELL SpMV formats, and obtained speedups of at least 10x concerning CPU-based kernels. Furthermore, we demonstrated via a performance characterization study that FPGAs outperform general-purpose processors in terms of compute time.
Zitieren
APA
Morales-Peña, A., & Meneses, E. (2024). Extending ginkgo to manage Reconfigurable Hardware-Based Kernels. Revista Colombiana de Computación, 25(2), 43-58. https://doi.org/10.29375/25392115.5276
Vancouver
Morales-Peña A, Meneses E. Extending ginkgo to manage Reconfigurable Hardware-Based Kernels. Revista Colombiana de Computación. 2024;25(2):43-58. doi:10.29375/25392115.5276
BibTeX
@article{MoralesPea2024,
author = {Morales-Peña, Alejandro and Meneses, Esteban},
title = {Extending ginkgo to manage Reconfigurable Hardware-Based Kernels},
journal = {Revista Colombiana de Computación},
year = {2024},
volume = {25},
number = {2},
pages = {43--58},
doi = {10.29375/25392115.5276}
}
EUVIMED