Abstract: An MPI Library implementing Direct Communication for Many-Core Based Accelerators.
Min SiYutaka IshikawaPublished in: SC Companion (2012)
Keyphrases
- parallel algorithm
- high level
- computing systems
- information systems
- communication systems
- general purpose
- message passing
- neural network
- parallel implementation
- parallel programming
- data sets
- communication overhead
- high performance computing
- massively parallel
- communication patterns
- parallelization strategy
- communication technologies
- communication channels
- communication cost
- efficient implementation
- computer networks
- data processing
- digital libraries
- multi agent