APR: A Novel Parallel Repacking Algorithm for Efficient GPGPU Parallel Code Transformation.
Yulong YuXubin HeHe GuoSihui ZhongYuxin WangXin ChenWeijun XiaoPublished in: GPGPU@ASPLOS (2014)
Keyphrases
- parallel implementation
- parallel version
- single pass
- objective function
- multiprocessor systems
- k means
- preprocessing
- worst case
- high accuracy
- optimization algorithm
- parallel processing
- detection algorithm
- parallel computation
- massively parallel
- high efficiency
- recognition algorithm
- learning algorithm
- simulated annealing
- computational cost
- experimental evaluation
- dynamic programming
- search space
- times faster
- computationally efficient
- expectation maximization
- parallel machines
- pruning strategy
- depth first search
- satisfiability testing