Dynamically Balanced Synchronization-Avoiding LU Factorization with Multicore and GPUs.
Simplice DonfackStanimire TomovJack J. DongarraPublished in: IPDPS Workshops (2014)
Keyphrases
- parallel programming
- graphics processing units
- multicore processors
- general purpose
- computing power
- shared memory
- matrix factorization
- low rank
- phase locked
- high end
- parallel processing
- real time
- computational power
- least squares
- pairwise
- highly parallel
- gpu implementation
- singular value decomposition
- parallel algorithm
- principal component analysis
- graphics hardware
- computer vision
- neural network
- data sets