Salvato in:
Dettagli Bibliografici
Autori principali: Moraru, Maxim, Kamalakkannan, Kamalavasan, Dominguez-Trujillo, Jered, Diehl, Patrick, Barai, Atanu, Loiseau, Julien, Baker, Zachary Kent, Pritchard, Howard, Shipman, Galen M
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2602.06085
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917252162387968
author Moraru, Maxim
Kamalakkannan, Kamalavasan
Dominguez-Trujillo, Jered
Diehl, Patrick
Barai, Atanu
Loiseau, Julien
Baker, Zachary Kent
Pritchard, Howard
Shipman, Galen M
author_facet Moraru, Maxim
Kamalakkannan, Kamalavasan
Dominguez-Trujillo, Jered
Diehl, Patrick
Barai, Atanu
Loiseau, Julien
Baker, Zachary Kent
Pritchard, Howard
Shipman, Galen M
contents FPGAs offer high performance, low latency, and energy efficiency for accelerated computing, yet adoption in scientific and edge settings is limited by the specialized hardware expertise required. High-level synthesis (HLS) boosts productivity over HDLs, but competitive designs still demand hardware-aware optimizations and careful dataflow design. We introduce LAAFD, an agentic workflow that uses large language models to translate general-purpose C++ into optimized Vitis HLS kernels. LAAFD automates key transfor mations: deep pipelining, vectorization, and dataflow partitioning and closes the loop with HLS co-simulation and synthesis feedback to verify correctness while iteratively improving execution time in cycles. Over a suite of 15 kernels representing common compute patterns in HPC, LAFFD achieves 99.9% geomean performance when compared to the hand tuned baseline for Vitis HLS. For stencil workloads, LAAFD matches the performance of SODA, a state-of-the-art DSL-based HLS code generator for stencil solvers, while yielding more readable kernels. These results suggest LAAFD substantially lowers the expertise barrier to FPGA acceleration without sacrificing efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2602_06085
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle LAAFD: LLM-based Agents for Accelerated FPGA Design
Moraru, Maxim
Kamalakkannan, Kamalavasan
Dominguez-Trujillo, Jered
Diehl, Patrick
Barai, Atanu
Loiseau, Julien
Baker, Zachary Kent
Pritchard, Howard
Shipman, Galen M
Distributed, Parallel, and Cluster Computing
FPGAs offer high performance, low latency, and energy efficiency for accelerated computing, yet adoption in scientific and edge settings is limited by the specialized hardware expertise required. High-level synthesis (HLS) boosts productivity over HDLs, but competitive designs still demand hardware-aware optimizations and careful dataflow design. We introduce LAAFD, an agentic workflow that uses large language models to translate general-purpose C++ into optimized Vitis HLS kernels. LAAFD automates key transfor mations: deep pipelining, vectorization, and dataflow partitioning and closes the loop with HLS co-simulation and synthesis feedback to verify correctness while iteratively improving execution time in cycles. Over a suite of 15 kernels representing common compute patterns in HPC, LAFFD achieves 99.9% geomean performance when compared to the hand tuned baseline for Vitis HLS. For stencil workloads, LAAFD matches the performance of SODA, a state-of-the-art DSL-based HLS code generator for stencil solvers, while yielding more readable kernels. These results suggest LAAFD substantially lowers the expertise barrier to FPGA acceleration without sacrificing efficiency.
title LAAFD: LLM-based Agents for Accelerated FPGA Design
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2602.06085