Characterizing numascale clusters with GPUs: MPI-based and GPU interconnect benchmarks.
Malik Muhammad Zaki Murtaza KhanAnne C. ElsterPublished in: HPCS (2016)
Keyphrases
- parallel programming
- graphics processing units
- parallel implementation
- parallel computing
- general purpose
- graphics hardware
- parallel algorithm
- gpu implementation
- parallel computation
- high performance computing
- clustering algorithm
- parallel architectures
- message passing interface
- graphics processors
- shared memory
- parallel processing
- high speed
- multi core processors
- massively parallel
- cpu implementation
- programming environment
- graphics cards
- real time
- cluster analysis
- fuzzy clustering
- heterogeneous computing
- message passing
- data clustering
- hierarchical clustering
- high end
- parallel computers
- computing systems
- parallel execution
- processing units
- cloud computing
- efficient implementation
- self organizing maps
- benchmark suite
- computational power
- compute unified device architecture
- commodity hardware
- gpu accelerated
- input data
- high dimensional