EUVIMEDEUVIMED entdecken
Crossref

FDRA: A Framework for a Dynamically Reconfigurable Accelerator Supporting Multi-Level Parallelism

Yunhui Qiu, Yiqing Mao, Xuchen Gao, Sichao Chen, Jiangnan Li, Wenbo Yin, Lingli Wang

ACM Transactions on Reconfigurable Technology and Systems · 31.03.2024 · 17 (1) · 1-26

DOI: 10.1145/3614224 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Reconfigurable Technology and Systems, online erschienen am 27.01.2024. Verfasst von Yunhui Qiu und 6 weiteren Autor:innen. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 27 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Coarse-grained reconfigurable architectures (CGRAs) have emerged as promising accelerators due to their high flexibility and energy efficiency. However, existing open source works often lack integration of CGRAs with CPU systems and corresponding toolchains. Moreover, there is rare support for the accelerator instruction pipelining to overlap data communication, computation, and configuration across multiple tasks. In this article, we propose FDRA, an open source exploration framework for a heterogeneous system-on-chip (SoC) with a RISC-V processor and a dynamically reconfigurable accelerator (DRA) supporting loop, instruction, and task levels of parallelism. FDRA encompasses parameterized SoC modeling, Verilog generation, source-to-source application code transformation using frontend and DRA compilers, SoC simulation, and FPGA prototyping. FDRA incorporates the extraction of periodic accumulative operators and multi-dimensional linear load/store operators from nested loops. The DRA enables accessing the shared L2 cache with virtual addresses and supports direct memory access with arbitrary start addresses and data lengths. Integrated into the RISC-V Rocket SoC, our DRA achieves a remarkable 55× acceleration for loop kernels and improves energy efficiency by 29×. Compared to state-of-the-art RISC-V vector units, our DRA demonstrates a 2.9× speed improvement and 3.5× greater energy efficiency. In contrast to previous CGRA+RISC-V SoCs, our SoC achieves a minimum speedup of 5.2×.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Qiu, Y., Mao, Y., Gao, X., Chen, S., Li, J., Yin, W., & Wang, L. (2024). FDRA: A Framework for a Dynamically Reconfigurable Accelerator Supporting Multi-Level Parallelism. ACM Transactions on Reconfigurable Technology and Systems, 17(1), 1-26. https://doi.org/10.1145/3614224

Vancouver

Qiu Y, Mao Y, Gao X, Chen S, Li J, Yin W, et al. FDRA: A Framework for a Dynamically Reconfigurable Accelerator Supporting Multi-Level Parallelism. ACM Transactions on Reconfigurable Technology and Systems. 2024;17(1):1-26. doi:10.1145/3614224

BibTeX
@article{Qiu2024,
  author = {Qiu, Yunhui and Mao, Yiqing and Gao, Xuchen and Chen, Sichao and Li, Jiangnan and Yin, Wenbo and Wang, Lingli},
  title = {FDRA: A Framework for a Dynamically Reconfigurable Accelerator Supporting Multi-Level Parallelism},
  journal = {ACM Transactions on Reconfigurable Technology and Systems},
  year = {2024},
  volume = {17},
  number = {1},
  pages = {1--26},
  doi = {10.1145/3614224}
}
Kostenlos starten →