FADO: Floorplan-Aware Directive Optimization Based on Synthesis and Analytical Models for High-Level Synthesis Designs on Multi-Die FPGAs
ACM Transactions on Reconfigurable Technology and Systems · 30.09.2024 · 17 (3) · 1-33
DOI: 10.1145/3653458 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Reconfigurable Technology and Systems, online erschienen am 17.09.2024. Verfasst von Linfeng Du und 8 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: RGC General Research Fund, AI Chip Center for Emerging Smart Systems (ACCESS), Hong Kong SAR.
Abstract (Originalsprache)
Multi-die FPGAs are widely adopted for large-scale accelerators, but optimizing high-level synthesis designs on these FPGAs faces two challenges. First, the delay caused by die-crossing nets creates an NP-hard floorplanning problem. Second, traditional directive optimization cannot consider resource constraints on each die or the timing issue incurred by the die-crossings. Furthermore, the high algorithmic complexity and the large scale lead to extended runtime for legalizing the floorplan of HLS designs under different directive configurations. To co-optimize the directives and floorplan of HLS designs on multi-die FPGAs, we formulate the co-search based on bin-packing variants and present two iterative optimization flows. The first (FADO 1.0) relies on a pre-built QoR library. It involves a greedy, latency-bottleneck-guided directive search, and an incremental floorplan legalization. Compared with a global floorplanning solution, it takes 693X~4925X shorter search time and achieves 1.16X~8.78X better design performance, measured in workload execution time. To remove the time-consuming QoR library generation, the second flow (FADO 2.0) integrates an analytical QoR model and redesigns the directive search to accelerate convergence. Through experiments on mixed dataflow and non-dataflow designs, compared with 1.0, FADO 2.0 further yields a 1.40X better design performance on average after implementation on the Alveo U250 FPGA.
Förderung
- RGC General Research Fund
- AI Chip Center for Emerging Smart Systems (ACCESS), Hong Kong SAR
Zitieren
APA
Du, L., Liang, T., Zhou, X., Ge, J., Li, S., Sinha, S., Zhao, J., Xie, Z., & Zhang, W. (2024). FADO: Floorplan-Aware Directive Optimization Based on Synthesis and Analytical Models for High-Level Synthesis Designs on Multi-Die FPGAs. ACM Transactions on Reconfigurable Technology and Systems, 17(3), 1-33. https://doi.org/10.1145/3653458
Vancouver
Du L, Liang T, Zhou X, Ge J, Li S, Sinha S, et al. FADO: Floorplan-Aware Directive Optimization Based on Synthesis and Analytical Models for High-Level Synthesis Designs on Multi-Die FPGAs. ACM Transactions on Reconfigurable Technology and Systems. 2024;17(3):1-33. doi:10.1145/3653458
BibTeX
@article{Du2024,
author = {Du, Linfeng and Liang, Tingyuan and Zhou, Xiaofeng and Ge, Jinming and Li, Shangkun and Sinha, Sharad and Zhao, Jieru and Xie, Zhiyao and Zhang, Wei},
title = {FADO: Floorplan-Aware Directive Optimization Based on Synthesis and Analytical Models for High-Level Synthesis Designs on Multi-Die FPGAs},
journal = {ACM Transactions on Reconfigurable Technology and Systems},
year = {2024},
volume = {17},
number = {3},
pages = {1--33},
doi = {10.1145/3653458}
}
EUVIMED