Contention-Aware Scheduling of Parallel Code for Heterogeneous Systems

A typical consumer desktop computer has a multi-core CPU with at least two and possibly up to eight processing ele- ments over two processors, and a multi-core GPU with up to 512 processing elements. Both the CPU and the GPU are ca- pable of running parallel code, yet it is not obvious when to utilize one processor or the other because of workload con- siderations and, as importantly, contention on each device. This paper demonstrates a method for dynamically deciding whether to run a given parallel workload on the CPU or the GPU depending on the state of the system when the code is aunched. To achieve this, we tested a selection of parallel penCL code on a multi-core CPU and a multi-core GPU, as part of a larger program that runs on the CPU. When the parallel code is launched, the runtime makes a dynamic deci- sion about which processor to run the code on, given system state and historical data. We demonstrate a method for using meta-data available to the runtime and historical data from code profiling to make the dynamic decision, and we out- line the runtime information necessary for making effective dynamic decisions, suggest hardware, operating system, and driver support.

Gregg, Chris, Jeff Brantley, and Kim Hazelwood. "Contention-Aware Scheduling of Parallel Code for Heterogeneous Systems." University of Virginia Dept. of Computer Science Tech Report (2010).