Abstract
This paper presents Heron, an FPGA-based special purpose processor core for pure, non-strict functional languages. We co-design its language semantics and parametrised design, gaining a high reductions-per-cycle performance metric. The Heron core is energy efficient, performing up to six times as many reductions per cycle as GHC. Despite its infancy, a 193 MHz Heron core outperforms wall-clock time for a mid-range Intel i3 1.9 GHz mobile CPU for 5 of these benchmarks and is competitive with an Alder Lake Intel i7 CPU. Its performance-per-Watt shows that the Heron core is a compelling solution for embedded applications. The simplicity of Heron's design results in just 2% FPGA resource usage, paving the way for future single-chip parallelism, further improving absolute performance.
Original language | English |
---|---|
Title of host publication | IFL '23: Proceedings of the 35th Symposium on Implementation and Application of Functional Languages |
Publisher | Association for Computing Machinery |
ISBN (Print) | 9798400716317 |
DOIs | |
Publication status | Published - 19 Jun 2024 |
Event | 35th Symposium on Implementation and Application of Functional Languages 2023 - Braga, Portugal Duration: 29 Aug 2023 → 31 Aug 2023 |
Conference
Conference | 35th Symposium on Implementation and Application of Functional Languages 2023 |
---|---|
Abbreviated title | IFL 2023 |
Country/Territory | Portugal |
City | Braga |
Period | 29/08/23 → 31/08/23 |
Keywords
- FPGAs
- functional languages
- graph reduction
- hardware accelerators
Fingerprint
Dive into the research topics of 'Heron: Modern Hardware Graph Reduction'. Together they form a unique fingerprint.Datasets
-
Dataset for results in "Heron: Modern Hardware Graph Reduction"
Ramsay, C. (Creator) & Stewart, R. J. (Supervisor), Heriot-Watt University, 23 Feb 2024
DOI: 10.17861/b9ab6ca4-a86c-4bf0-b6f3-e462129b6ebb
Dataset
-
Dataset for the Heron Core in "Heron: Modern Hardware Graph Reduction"
Ramsay, C. (Creator) & Stewart, R. J. (Supervisor), Heriot-Watt University, 21 Feb 2024
DOI: 10.17861/f4fab5ef-ae98-4300-8328-ea59e47ff8c6, https://github.com/haflang/heron
Dataset