Benchmarking the Nvidia GPU Lineage: From Early K80 to Modern A100 with Asynchronous Memory Transfers.
Martin SvedinSteven Wei Der ChienGibson ChikafaNiclas JanssonArtur PodobasPublished in: HEART (2021)
Keyphrases
- graphics processors
- graphics processing units
- parallel implementation
- graphics hardware
- gpu implementation
- general purpose
- compute unified device architecture
- memory usage
- parallel computing
- cpu implementation
- real time
- memory requirements
- memory space
- parallel algorithm
- computing power
- times faster
- metadata
- databases
- scientific data
- data quality
- data transfer
- main memory
- random access
- limited memory
- multi threaded
- memory size
- data structure