Litcius/Paper detail

Graphite

Zhangxiaowen Gong, Houxiang Ji, Yao Yao, Christopher W. Fletcher, Christopher J. Hughes, Josep Torrellas

202226 citationsDOI

Abstract

Graph Neural Networks (GNNs) are becoming popular because they are effective at extracting information from graphs. To execute GNNs, CPUs are good platforms because of their high availability and terabyte-level memory capacity, which enables full-batch computation on large graphs. However, GNNs on CPUs are heavily memory bound, which limits their performance.

Topics & Concepts

TerabyteComputer scienceParallel computingComputationGraphTheoretical computer scienceAlgorithmOperating systemAdvanced Graph Neural NetworksGraph Theory and AlgorithmsMachine Learning in Materials Science