Enhancing MPI performance using atomic pipelined message broadcast in a distributed memory MPSoC.
Jongsu ParkHeejun YunSangook MoonPublished in: IEICE Electron. Express (2014)
Keyphrases
- distributed memory
- parallel architecture
- shared memory
- parallel implementation
- multiprocessor systems
- scientific computing
- ibm sp
- parallel computers
- fine grain
- parallel algorithm
- message passing
- data parallelism
- single processor
- parallel machines
- multi processor
- matrix multiplication
- multithreading
- parallel computation
- parallel computing
- computer vision
- data flow
- belief propagation
- pairwise
- lower bound
- information systems