MGG: Accelerating Graph Neural Networks with Fine-Grained Intra-Kernel Communication-Computation Pipelining on Multi-GPU Platforms.
Yuke WangBoyuan FengZheng WangTong GengKevin J. BarkerAng LiYufei DingPublished in: OSDI (2023)
Keyphrases
- fine grained
- neural network
- coarse grained
- parallel computation
- real time
- messaging service
- access control
- graph kernels
- tightly coupled
- parallel processing
- kernel methods
- parallel computing
- artificial neural networks
- kernel function
- information extraction
- weighted graph
- machine learning
- data lineage
- databases
- parallel implementation
- data model
- support vector