OpenPARF: An Open-Source Placement and Routing Framework for Large-Scale Heterogeneous FPGAs with Deep Learning Toolkit

TL;DR

OpenPARF is an open-source FPGA placement and routing framework leveraging deep learning and GPU parallelization, achieving improved wirelength and faster placement for large-scale designs.

Contribution

It introduces a novel asymmetric electrostatic field system and supports large-scale irregular routing, advancing FPGA physical design automation.

Findings

Achieves 0.4-12.7% reduction in routed wirelength.

More than 2x speedup in placement time.

Supports large-scale irregular routing resource graphs.

Abstract

This paper proposes OpenPARF, an open-source placement and routing framework for large-scale FPGA designs. OpenPARF is implemented with the deep learning toolkit PyTorch and supports massive parallelization on GPU. The framework proposes a novel asymmetric multi-electrostatic field system to solve FPGA placement. It considers fine-grained routing resources inside configurable logic blocks (CLBs) for FPGA routing and supports large-scale irregular routing resource graphs. Experimental results on ISPD 2016 and ISPD 2017 FPGA contest benchmarks and industrial benchmarks demonstrate that OpenPARF can achieve 0.4-12.7% improvement in routed wirelength and more than $2\times$ speedup in placement. We believe that OpenPARF can pave the road for developing FPGA physical design engines and stimulate further research on related topics.