GaaS workload characterization under NUMA architecture for virtualized GPU.
Huixiang ChenMeng WangYang HuMingcong SongTao LiPublished in: ISPASS (2017)
Keyphrases
- real time
- execution model
- management system
- load balancing
- heterogeneous computing
- response time
- memory access
- virtual machine
- message passing
- data center
- software architecture
- parallel processing
- resource management
- network architecture
- shared memory
- graphics hardware
- multithreading
- cloud computing
- computation intensive
- general purpose
- graphic processing unit