OpenMP to CUDA graphs: a compiler-based transformation to enhance the programmability of NVIDIA devices.
Chenle YuSara RoyuelaEduardo QuiñonesPublished in: SCOPES (2020)
Keyphrases
- parallel programming
- compute unified device architecture
- shared memory
- parallel implementation
- graphics processing units
- general purpose
- parallel computing
- parallel algorithm
- gpu implementation
- graphics processors
- parallel execution
- high end
- parallel computation
- graphics hardware
- multi core processors
- programming language
- graph mining
- graph matching
- mobile devices
- processing units
- message passing
- parallel processing
- smart phones
- massively parallel
- weighted graph
- bipartite graph
- programming environment
- times faster
- general purpose computing
- highly optimized
- distributed memory machines
- parallel architectures
- distributed memory
- high performance computing
- graph theoretic
- undirected graph
- directed graph