Designing Efficient Index-Digit Algorithms for CUDA GPU Architectures.
Jacobo LobeirasMargarita AmorRamon DoalloPublished in: IEEE Trans. Parallel Distributed Syst. (2016)
Keyphrases
- parallel architectures
- significant improvement
- graphics processors
- computationally efficient
- computationally complex
- computationally intensive
- orders of magnitude
- highly scalable
- general purpose
- real time
- indexing structure
- parallel implementation
- times faster
- computationally expensive
- computationally demanding
- theoretical analysis
- gpu accelerated
- data structure
- algorithmic solutions
- b tree
- complexity analysis
- parallel computing
- efficient implementation
- benchmark datasets
- optimization problems
- computational cost
- learning algorithm
- machine learning