Accelerating LINPACK with MPI-OpenCL on Clusters of Multi-GPU Nodes.
Gangwon JoJeongho NahJun LeeJungwon KimJaejin LeePublished in: IEEE Trans. Parallel Distributed Syst. (2015)
Keyphrases
- high performance computing
- parallel programming
- graphics processing units
- parallel computing
- message passing interface
- parallel implementation
- parallel algorithm
- shared memory
- massively parallel
- general purpose
- parallel architectures
- parallel computation
- real time
- parallel processing
- processing units
- clustering algorithm
- computing systems
- gpu implementation
- data objects
- cluster analysis
- data points
- computing resources
- fuzzy c means
- hierarchical clustering
- cohesive subgroups
- nodes of a graph
- multithreading
- shortest path
- graphics hardware
- fine grained
- message passing
- graph structure
- energy efficiency
- grid computing