Enabling GPU Memory Oversubscription via Transparent Paging to an NVMe SSD.
Joshua BakitaJames H. AndersonPublished in: RTSS (2022)
Keyphrases
- virtual memory
- main memory
- graphics processors
- real time
- operating system
- memory requirements
- multi threaded
- graphics hardware
- memory usage
- parallel computing
- parallel implementation
- gpu accelerated
- compute intensive
- memory space
- solid state
- graphics processing units
- parallel computation
- database
- gpu implementation
- hard disk
- memory bandwidth
- heterogeneous computing