A CUDA-MPI Hybrid Bitonic Sorting Algorithm for GPU Clusters.
Sam WhiteNiels J. VeroskyTia NewhallPublished in: ICPP Workshops (2012)
Keyphrases
- parallel implementation
- objective function
- gpu implementation
- hierarchical clustering
- cost function
- times faster
- dynamic programming
- detection algorithm
- k means
- learning algorithm
- optimization algorithm
- matching algorithm
- computational cost
- computational complexity
- general purpose
- worst case
- np hard
- search space
- preprocessing
- parallel computing
- parallel computation
- arbitrary shaped
- expectation maximization
- probabilistic model
- data clustering
- massively parallel
- gpu accelerated
- sorting algorithms