SnuRHAC: A Runtime for Heterogeneous Accelerator Clusters with CUDA Unified Memory.
Jaehoon JungDaeyoung ParkGangwon JoJungho ParkJaejin LeePublished in: HPDC (2021)
Keyphrases
- parallel implementation
- clustering algorithm
- compute intensive
- general purpose
- hierarchical clustering
- cluster analysis
- memory usage
- memory requirements
- memory space
- self organizing maps
- real world
- neural network
- efficient implementation
- fuzzy clustering
- computing power
- parallel computation
- limited memory
- document clustering
- data clustering
- hierarchical structure
- subspace clustering
- data points
- data mining
- unified model
- resource consumption