CODO: An Automated Compiler for Comprehensive Dataflow Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Weichuang, Wang, Yiquan, Zhang, Xinzhou, Zhang, Chi, Feng, Yu, Hou, Xiaofeng, Li, Chao, Zhao, Jieru, Guo, Minyi
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914471291650048
author Zhang, Weichuang
Wang, Yiquan
Zhang, Xinzhou
Zhang, Chi
Feng, Yu
Hou, Xiaofeng
Li, Chao
Zhao, Jieru
Guo, Minyi
author_facet Zhang, Weichuang
Wang, Yiquan
Zhang, Xinzhou
Zhang, Chi
Feng, Yu
Hou, Xiaofeng
Li, Chao
Zhao, Jieru
Guo, Minyi
contents FPGAs are well-suited for dataflow architectures that process data in a streaming or pipelined manner, thus satisfying the high computational and communication demands of emerging applications. However, manually implementing an efficient dataflow architecture for large-scale applications is still challenging, even for specialists who use high-level synthesis (HLS) to simplify FPGA programming. To address this, we introduce CODO, an automated compiler that generates feasible and efficient dataflow accelerators on FPGAs. CODO features a systematic method for detecting and eliminating both coarse-grained and fine-grained dataflow violations. Building on this, CODO performs both on- and off-chip data movement optimizations to maximize transfer efficiency. To guarantee a higher design quality, CODO performs automatic scheduling to generate high-performance dataflow accelerators, ensuring a balanced performance-resource trade-off. Synthesis results show that CODO delivers $1.45\times$ to $4.52\times$ latency speedups on typical computation kernels and $3.7\times$ to $33.8\times$ speedups on DNN models compared to SOTA frameworks. In on-board evaluations, CODO achieves $7.3\times$ average speedup on CNN models and $2.07\times$ average speedup on the GPT-2 model over SOTA frameworks. The compiler is open-sourced at https://github.com/sjtu-zhao-lab/codo-artifact.
format Preprint
id arxiv_https___arxiv_org_abs_2604_12618
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CODO: An Automated Compiler for Comprehensive Dataflow Optimization
Zhang, Weichuang
Wang, Yiquan
Zhang, Xinzhou
Zhang, Chi
Feng, Yu
Hou, Xiaofeng
Li, Chao
Zhao, Jieru
Guo, Minyi
Hardware Architecture
FPGAs are well-suited for dataflow architectures that process data in a streaming or pipelined manner, thus satisfying the high computational and communication demands of emerging applications. However, manually implementing an efficient dataflow architecture for large-scale applications is still challenging, even for specialists who use high-level synthesis (HLS) to simplify FPGA programming. To address this, we introduce CODO, an automated compiler that generates feasible and efficient dataflow accelerators on FPGAs. CODO features a systematic method for detecting and eliminating both coarse-grained and fine-grained dataflow violations. Building on this, CODO performs both on- and off-chip data movement optimizations to maximize transfer efficiency. To guarantee a higher design quality, CODO performs automatic scheduling to generate high-performance dataflow accelerators, ensuring a balanced performance-resource trade-off. Synthesis results show that CODO delivers $1.45\times$ to $4.52\times$ latency speedups on typical computation kernels and $3.7\times$ to $33.8\times$ speedups on DNN models compared to SOTA frameworks. In on-board evaluations, CODO achieves $7.3\times$ average speedup on CNN models and $2.07\times$ average speedup on the GPT-2 model over SOTA frameworks. The compiler is open-sourced at https://github.com/sjtu-zhao-lab/codo-artifact.
title CODO: An Automated Compiler for Comprehensive Dataflow Optimization
topic Hardware Architecture
url https://arxiv.org/abs/2604.12618