Optimized Non-contiguous MPI Datatype Communication for GPU Clusters: Design, Implementation and Evaluation with MVAPICH2.
Hao WangSreeram PotluriMiao LuoAshish Kumar SinghXiangyong OuyangSayantan SurDhabaleswar K. PandaPublished in: CLUSTER (2011)
Keyphrases
- parallel implementation
- efficient implementation
- implementation issues
- parallel distributed
- formative evaluation
- parallel algorithm
- multimedia communication
- communication systems
- architectural design
- future development
- circuit design
- pilot testing
- real time
- cluster of workstations
- active participation
- hardware design
- parallel programming
- parallel computation
- parallel computing
- design methodology
- message passing
- design principles
- cluster analysis
- information sharing
- graphical models
- clustering algorithm