Back to HomeBack to News

One paper accepted by Euro-Par 2024

2024-05-01

The paper titled “GDL-GNN: Applying GPU Dataloading of Large Datasets for Graph Neural Network Inference” is accepted by the 30th International European Conference on Parallel and Distributed Computing (Euro-Par 2024). This paper observes the possibility to load the graph data directly from GPU memory on the basis of graph partitioning, along with several optimizations to further enhance the speed.

Haoran is the first author of the paper. Euro-Par 2024 will take place in Madrid, Spain on 26-30 Aug 2024. Acceptance rate this year is 30%.

[Update 2024-08-29] Haoran gave a presentation of this paper in Euro-Par 2024 at UC3M.


Tags: ICT, GIMLab, Paper