Hybrid Static/Dynamic Schedules for Tiled Polyhedral Programs

TL;DR

This paper introduces a hybrid static/dynamic scheduling system for tiled polyhedral programs, reducing overheads and energy consumption while supporting various parallel computing platforms.

Contribution

It presents a novel hybrid scheduling approach that combines static code structure with dynamic enforcement, improving efficiency over previous methods.

Findings

Overheads are reduced by at least one polynomial degree compared to prior techniques.

The system is adaptable to multiple parallel computing platforms including GPUs and MPI.

Experiments show 24% to 70% energy savings with the proposed mechanism.

Abstract

Polyhedral compilers perform optimizations such as tiling and parallelization; when doing both, they usually generate code that executes "barrier-synchronized wavefronts" of tiles. We present a system to express and generate code for hybrid schedules, where some constraints are automatically satisfied through the structure of the code, and the remainder are dynamically enforced at run-time with data flow mechanisms. We prove bounds on the added overheads that are better, by at least one polynomial degree, than those of previous techniques. We propose a generic mechanism to implement the needed synchronization, and show it can be easily realized for a variety of targets: OpenMP, Pthreads, GPU (CUDA or OpenCL) code, languages like X10, Habanero, Cilk, as well as data flow platforms like DAGuE, and OpenStream and MPI. We also provide a simple concrete implementation that works without the need of any sophisticated run-time mechanism. Our experiments show our simple implementation to be competitive or better than the wavefront-synchronized code generated by other systems. We also show how the proposed mechanism can achieve 24% to 70% reduction in energy.