Multithreaded double queuing for balanced CPU-GPU memory copying.
Sanghun ChoJaewan HongJungsik ChoiHwansoo HanPublished in: SAC (2019)
Keyphrases
- multithreading
- memory bandwidth
- parallel computing
- multi threaded
- parallel programming
- graphics processors
- intel xeon
- computational power
- graphics processing units
- highly efficient
- shared memory
- memory access
- gpu implementation
- end to end
- queuing model
- cache misses
- distributed memory
- limited memory
- general purpose
- heterogeneous computing
- multi core processors
- neural network
- data transfer
- round robin
- memory efficient
- processing units
- coarse grained
- computer architecture
- parallel algorithm
- memory hierarchy
- message passing
- real time
- parallel computation
- secondary storage
- compute unified device architecture