Guided Region-Based GPU Scheduling: Utilizing Multi-thread Parallelism to Hide Memory Latency.
Jianmin ChenXi TaoZhen YangJih-Kwon PeirXiaoyuan LiShih-Lien LuPublished in: IPDPS (2013)
Keyphrases
- memory bandwidth
- level parallelism
- heterogeneous computing
- processing power
- compute intensive
- parallel programming
- floating point
- processing units
- commodity hardware
- scheduling problem
- scheduling algorithm
- memory access
- data transfer
- parallel computers
- parallel computing
- parallel processing
- cache misses
- computing power
- parallel computation
- parallel machines
- real time database systems
- single instruction multiple data
- response time
- parallel algorithm
- multiprocessor systems
- multithreading
- round robin
- graphics processors
- memory requirements
- resource consumption
- general purpose
- resource utilization
- data structure