IMA-GNN: In-Memory Acceleration of Centralized and Decentralized Graph Neural Networks at the Edge

Mehrdad Morsali, Mahmoud Nazzal, Abdallah Khreishah, Shaahin Angizi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

In this paper, we propose IMA-GNN as an In-Memory Accelerator for centralized and decentralized Graph Neural Network inference, explore its potential in both settings and provide a guideline for the community targeting flexible and efficient edge computation. Leveraging IMA-GNN, we first model the computation and communication latencies of edge devices. We then present practical case studies on GNN-based taxi demand and supply prediction and also adopt four large graph datasets to quantitatively compare and analyze centralized and decentralized settings. Our cross-layer simulation results demonstrate that on average, IMA-GNN in the centralized setting can obtain ∼790x communication speed-up compared to the decentralized GNN setting. However, the decentralized setting performs computation ∼1400x faster while reducing the power consumption per device. This further underlines the need for a hybrid semi-decentralized GNN approach.

Original languageEnglish (US)
Title of host publicationGLSVLSI 2023 - Proceedings of the Great Lakes Symposium on VLSI 2023
PublisherAssociation for Computing Machinery
Pages3-8
Number of pages6
ISBN (Electronic)9798400701252
DOIs
StatePublished - Jun 5 2023
Event33rd Great Lakes Symposium on VLSI, GLSVLSI 2023 - Knoxville, United States
Duration: Jun 5 2023Jun 7 2023

Publication series

NameProceedings of the ACM Great Lakes Symposium on VLSI, GLSVLSI

Conference

Conference33rd Great Lakes Symposium on VLSI, GLSVLSI 2023
Country/TerritoryUnited States
CityKnoxville
Period6/5/236/7/23

All Science Journal Classification (ASJC) codes

  • General Engineering

Keywords

  • edge computing
  • graph neural network
  • in-memory computing

Fingerprint

Dive into the research topics of 'IMA-GNN: In-Memory Acceleration of Centralized and Decentralized Graph Neural Networks at the Edge'. Together they form a unique fingerprint.

Cite this