Design flows use graph partitioning both as a precursor to place and route for single devices, and to divide netlists or task graphs among multiple devices. Partitioners have accommodated FPGA heterogeneity via multi-resource constraints, but have not yet exploited the corresponding ability to implement some computations in multiple ways (e.g., LUTs vs. DSP blocks), which could enable a superior solution. This paper introduces multi-personality graph partitioning, which incorporates aspects of resource mapping into partitioning. We present a modified multi-level KLFM partitioning algorithm that also performs heterogeneous resource mapping for nodes with multiple potential implementations (multiple personalities). We evaluate several variants of our multi-personality FPGA circuit partitioner using 21 circuits and benchmark graphs, and show that dynamic resource mapping improves cut size on average by 27% over static mapping for these circuits. We further show that it improves deviation from target resource utilizations by 50% over post-partitioning resource mapping.
@article{arxiv.1704.01676,
title = {Multi-Personality Partitioning for Heterogeneous Systems},
author = {Anthony Gregerson and Aman Chadha and Katherine Morrow},
journal= {arXiv preprint arXiv:1704.01676},
year = {2017}
}
Comments
International Conference on Field-Programmable Technology (ICFPT), Kyoto Research Park, Japan, Dec. 9-11, 2013. hardware design; hardware architecture; cad; computer aided design; IC design; integrated circuit design; partitioning algorithms; field programmable gate arrays; benchmark testing; heuristic algorithms; resource management; dynamic scheduling; digital signal processing