BitGNN: Unleashing the Performance Potential of Binary Graph Neural Networks on GPUs

Jou-An Chen, Hsin-Hsuan Sung, Xipeng Shen, Sutanay Choudhury, Ang Li
5/4/2023
cs.DCcs.LG

No confident code implementation found for this paper.

Abstract

Recent studies have shown that Binary Graph Neural Networks (GNNs) are promising for saving computations of GNNs through binarized tensors. Prior work, however, mainly focused on algorithm designs or training techniques, leaving it open to how to materialize the performance potential on accelerator hardware fully. This work redesigns the binary GNN inference backend from the efficiency perspective. It fills the gap by proposing a series of abstractions and techniques to map binary GNNs and their computations best to fit the nature of bit manipulations on GPUs. Results on real-world graphs with GCNs, GraphSAGE, and GraphSAINT show that the proposed techniques outperform state-of-the-art binary GNN implementations by 8-22X with the same accuracy maintained. BitGNN code is publicly available.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Single-GPU GNN Systems: Traps and Pitfalls

83% match2024

AdaptGear: Accelerating GNN Training via Adaptive Subgraph-Level Kernels on GPUs

83% match2023

GNNHLS: Evaluating Graph Neural Network Inference via High-Level Synthesis

83% match1 repo122023

A Survey on Graph Neural Network Acceleration: Algorithms, Systems, and Customized Hardware

83% match2023

GNNBuilder: An Automated Framework for Generic Graph Neural Network Accelerator Generation, Simulation, and Optimization

83% match1 repo252023

Characterizing and Understanding HGNN Training on GPUs

83% match2024

Cite this paper

@article{chen2023bitgnn,
  title  = {BitGNN: Unleashing the Performance Potential of Binary Graph Neural Networks on GPUs},
  author = {Jou-An Chen and Hsin-Hsuan Sung and Xipeng Shen and Sutanay Choudhury and Ang Li},
  year   = {2023},
  eprint = {2305.02522},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.02522v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.02522.svg)](https://codeofpaper.com/paper/2305.02522)

Discussion