high performance computing on graphics processing units: hgpu.org

General-purpose graphical processing units (GPGPUs) have transformed high-performance computing over the past decade. Making great computational power available with reduced cost and power consumption overheads, heterogeneous CPU-GPU-equipped systems have helped to make possible the emerging class of exascale data-intensive applications. Although the theoretical performance achieved by these hybrid systems is impressive, taking practical advantage of this computing power remains a very challenging problem. Most applications are still being deployed to either GPU or CPU, leaving the other resource under or un-utilized. In this paper, we describe techniques for dynamic smart work partitioning, load balancing, and performance-aware task grouping in order to make efficient collaborative use of available CPUs and GPUs. In the context of a largescale image analysis application, our evaluations show that intelligently co-scheduling CPUs and GPUs can significantly improve performance over GPU-only or multi-core CPU-only approaches.