DGI: An Easy and Efficient Framework for GNN Model Evaluation

Peiqi Yin; Xiao Yan; Jinjing Zhou; Qiang Fu; Zhenkun Cai; James Cheng; Bo Tang; Minjie Wang

10.1145/3580305.3599805

2023

29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)

PROCEEDINGS OF THE 29TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2023

AUG 06-10, 2023

null,Long Beach,CA

1601 Broadway, 10th Floor, NEW YORK, NY, UNITED STATES

ASSOC COMPUTING MACHINERY

["While many systems have been developed to train graph neural networks (GNNs), efficient model evaluation, which computes node embedding according to a given model, remains to be addressed. For instance, using the widely adopted node-wise approach, model evaluation can account for over 90% of the time in the end-to-end training process due to neighbor explosion, which means that a node accesses its multi-hop neighbors. The layer-wise approach avoids neighbor explosion by conducting computation layer by layer in GNN models. However, layer-wise model evaluation takes considerable implementation efforts because users need to manually decompose the GNN model into layers, and different implementations are required for GNN models with different structures.","In this paper, we present DGI-a framework for easy and efficient GNN model evaluation, which automatically translates the training code of a GNN model for layer-wise evaluation to minimize user effort. DGI is general for different GNN models and evaluation requests (e.g., computing embedding for all or some of the nodes), and supports out-of-core execution on large graphs that cannot fit in CPU memory. Under the hood, DGI traces the computation graph of GNN model, partitions the computation graph into layers that are suitable for layer-wise evaluation according to tailored rules, and executes each layer efficiently by reordering the computation tasks and managing device memory consumption. Experiment results show that DGI matches hand-written implementations of layer-wise evaluation in efficiency and consistently outperforms node-wise evaluation across different datasets and hardware settings, and the speedup can be over 1,000x."]

Graph neural networks System for machine learning

其他

英语

EI ; CPCI-S

CUHK direct grant[4055146] ; Guangdong Basic and Applied Basic Research Foundation[2021A1515110067] ; Shenzhen Fundamental Research Program[20220815112848002]

Computer Science

Computer Science, Information Systems ; Computer Science, Interdisciplinary Applications ; Computer Science, Theory & Methods

WOS:001118896305043

Web of Science

1.AWS Shanghai AI Lab
2.Southern University of Science and Technology
3.TensorChord
4.George Washington University
5.The Chinese University of Hong Kong

Peiqi Yin,Xiao Yan,Jinjing Zhou,et al. DGI: An Easy and Efficient Framework for GNN Model Evaluation[C]. 1601 Broadway, 10th Floor, NEW YORK, NY, UNITED STATES:ASSOC COMPUTING MACHINERY,2023.