Offloading Accelerator-intensive Workloads in CPU-GPU Heterogeneous Processors.
Nandinbaatar TsogSaad MubeenFredrik BruhnMoris BehnamMikael SjödinPublished in: ETFA (2021)
Keyphrases
- heterogeneous computing
- compute intensive
- graphics processing units
- parallel implementation
- multithreading
- parallel computation
- graphics processors
- parallel processing
- parallel processors
- parallel computing
- parallel algorithm
- general purpose
- gpu implementation
- parallel computers
- graphics hardware
- distributed memory
- parallel programming
- message passing interface
- high performance computing
- computer architecture
- real time
- computing systems
- memory bandwidth
- high end
- computer systems
- highly parallel
- embedded processors
- single processor
- cpu implementation
- compute unified device architecture
- memory hierarchy
- data transfer
- field programmable gate array
- efficient implementation
- parallel architectures
- shared memory
- massively parallel
- database systems
- memory access
- data access
- pc cluster
- grid computing
- data processing
- floating point
- single instruction multiple data