Improving Performance of the Hypre Iterative Solver for Uintah Combustion Codes on Manycore Architectures Using MPI Endpoints and Kernel Consolidation.
Damodar SahasrabudheMartin BerzinsPublished in: ICCS (1) (2020)
Keyphrases
- endpoints
- parallel architectures
- message passing interface
- shared memory
- minimal cost
- kernel function
- error correction
- message passing
- parallel computing
- parallel processing
- parallel algorithm
- parallel computers
- kernel methods
- parallel implementation
- sensitivity analysis
- minimal path
- general purpose
- parallelization strategy
- diesel engine
- neural network
- continuous functions
- distributed memory
- massively parallel
- support vector
- positive definite
- multiple kernel learning
- medial axis