High-performance Cholesky factorization for GPU-only execution.
Azzam HaidarAhmad AbdelfattahStanimire TomovJack J. DongarraPublished in: GPGPU@PPoPP (2017)
Keyphrases
- graphics processing units
- matrix factorization
- graphics hardware
- parallel computing
- pairwise
- execution model
- real time
- singular value decomposition
- parallel programming
- kronecker product
- multithreading
- parallel computation
- parallel processing
- data flow
- parallel execution
- low rank
- general purpose
- high reliability
- data intensive
- highly parallel
- gpu implementation
- graphics processors