After clicking “Watch Now” you will be prompted to login or join.
Accelerating DNN Inference with GraphBLAS and the GPU
Xiaoyun Wang, University of California, Davis
GTC 2020
Our work addresses the 2019 Sparse Deep Neural Network Graph Challenge with an implementation using the GraphBLAS programming model. We'll demonstrate our solution to this challenge with GraphBLAST, a GraphBLAS implementation on the GPU, and compare it to SuiteSparse, a GraphBLAS implementation on the CPU. The GraphBLAST implementation is 1.94x faster than SuiteSparse; the primary opportunity to increase performance on the GPU is a higher-performance sparse-matrix-times-sparse-matrix (SpGEMM) kernel.