Bolt: Bridging the Gap between Auto-tuners and Hardware-native Performance.
Jiarong XingLeyuan WangShang ZhangJack ChenAng ChenYibo ZhuPublished in: MLSys (2022)
Keyphrases
- low cost
- hardware and software
- computing power
- real time
- hardware implementation
- image processing
- computer systems
- computing systems
- software implementation
- high end
- hardware design
- field programmable gate array
- general purpose
- multiscale
- massively parallel
- efficient implementation
- data acquisition
- parallel architectures
- data sets