Evolving Cut-Off Mechanisms and Other Work-Stealing Parameters for Parallel Programs

Optimizing parallel programs is a complex task because the interference among many different parameters. Work-stealing runtimes, used to dynamically balance load among different processor cores, are no exception. This work explores the automatic configuration of the following runtime parameters: dynamic granularity control algorithms, granularity control cache, work-stealing algorithm, lazy binary splitting parameter, the maximum queue size and the unparking interval. The performance of the program is highly sensible to the granularity control algorithm, which can be a combination of other granularity algorithms. In this work, we address two search-based problems: finding a globally efficient work-stealing configuration, and finding the best configuration just for an individual program. For both problems, we propose the use of a Genetic Algorithm (GA). The genotype of the GA is able to represent combinations of up to three cut-off algorithms, as well as other work-stealing parameters. The proposed GA has been evaluated in its ability to obtain a more efficient solution across a set of programs, in its ability to generalize the solution to a larger set of programs, and its ability to evolve single programs individually. The GA was able to improve the performance of the set of programs in the training set, but the obtained configurations were not generalized to a larger benchmark set. However, it was able to successfully improve the performance of each program individually.