Enhanced GPU Resource Utilization through Fairness-aware Task Scheduling

Underutilization as well as oversubscription of processing resources are common problems in current accelerator-based computing systems. Facing these challenges will require intelligent algorithms for scheduling parallel workloads on accelerators. The general aim of this paper is to achieve fair distribution of the tremendous computation power of modern devices among running applications towards enhancing resource utilization. Given a set of real applications, we evaluate our model and explore the advantages of multi-tasking and concurrency on current GPUs.