Source code for "HP-GNN: Generating High Throughput GNN Training Implementation on CPU-FPGA Heterogeneous Platform", ACM/SIGDA FPGA 2022.
File structure:
HP-GNN/
|
│ README.md
|
└───/src #kernel designs and some supportive functions
│ │ SPMM kernel
│ │ GEMM kernel
| | Design space exploration (DSE) engine
│ │ ...
└─
Citation
@inproceedings{hp-gnn,
author = {Lin, Yi-Chien and Zhang, Bingyi and Prasanna, Viktor},
title = {HP-GNN: Generating High Throughput GNN Training Implementation on CPU-FPGA Heterogeneous Platform},
booktitle = {Proceedings of the 2022 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays},
year = {2022}
}