Persistent Kernels for Iterative Memory-bound GPU Applications.
Lingqi ZhangMohamed WahibPeng ChenJintao MengXiao WangSatoshi MatsuokaPublished in: CoRR (2022)
Keyphrases
- real time
- lower bound
- upper bound
- intel xeon
- graphics hardware
- graphics processors
- memory space
- error bounds
- multi threaded
- graphics processing units
- memory requirements
- support vector
- main memory
- gaussian processes
- parallel implementation
- general purpose
- worst case
- memory usage
- data transfer
- feature space
- parallel computation
- gpu implementation
- memory size
- gpu accelerated
- parallel computing
- graph kernels
- kernel learning
- computational power
- database systems
- machine learning