High Performance and Enhanced Scalability for Parallel Applications using MPI-3's non-blocking Collectives.
Surendra Varma PericherlaSathish VadhiyarPublished in: ICCS (2017)
Keyphrases
- distributed memory
- pc cluster
- shared memory
- parallel implementation
- parallel algorithm
- parallel computers
- message passing interface
- parallel programming
- parallel computing
- parallelization strategy
- multithreading
- ibm sp
- parallel processing
- massively parallel
- message passing
- distributed memory machines
- scientific computing
- highly parallel
- parallel architectures
- computer architecture
- high performance computing
- coarse grained
- high efficiency
- graphics processing units
- parallel architecture
- parallel machines
- fault tolerance
- multi core processors
- parallel computation
- scalability issues
- parallel execution
- collective intelligence
- load balance
- general purpose
- parallel database systems
- high reliability
- data intensive computing
- compute intensive
- data partitioning
- data sets
- response time
- neural network