Interleaved Parallel Schemes: Improving Memory Throughput on Supercomputers.
André SeznecJacques LenfantPublished in: ISCA (1992)
Keyphrases
- parallel programming
- memory bandwidth
- shared memory
- massively parallel
- parallel architectures
- single processor
- parallel computing
- distributed memory
- parallel hardware
- memory footprint
- allocation scheme
- processing elements
- response time
- model based predictive control
- level parallelism
- multi core processors
- memory requirements
- high performance computing
- parallel processing
- neural network
- multicore processors
- commodity hardware
- computing power
- high end
- memory space
- resource utilization
- data transfer
- memory size
- data structure
- processing units
- memory usage
- computational power
- limited memory
- computer architecture
- main memory
- message passing interface
- load balancing
- energy consumption
- parallel computation
- distributed shared memory