Architectural Implications of GNN Aggregation Programming Abstractions
This work addresses the need for efficient GNN computation by evaluating abstractions for researchers and practitioners, but it is incremental as it focuses on analysis rather than new methods.
The authors tackled the lack of comprehensive evaluation of programming abstractions for GNN aggregation by classifying them based on data organization and propagation methods, and through a characterization study on a state-of-the-art GNN library, they provided performance comparisons and insights for future acceleration.
Graph neural networks (GNNs) have gained significant popularity due to the powerful capability to extract useful representations from graph data. As the need for efficient GNN computation intensifies, a variety of programming abstractions designed for optimizing GNN Aggregation have emerged to facilitate acceleration. However, there is no comprehensive evaluation and analysis upon existing abstractions, thus no clear consensus on which approach is better. In this letter, we classify existing programming abstractions for GNN Aggregation by the dimension of data organization and propagation method. By constructing these abstractions on a state-of-the-art GNN library, we perform a thorough and detailed characterization study to compare their performance and efficiency, and provide several insights on future GNN acceleration based on our analysis.