On the Duality of Task and Actor Programming Models

TL;DR

This paper reveals a duality between task and actor programming models, demonstrating how task-based systems can achieve actor-like performance with enhanced productivity through specific techniques.

Contribution

It introduces techniques that bridge the performance gap between task and actor models, enabling task systems to match actor performance without sacrificing productivity.

Findings

Realm's overheads reduced by 1.7-5.3x

Legion's strong scaling improved by 1.3-5.0x

Task models can match actor performance with new techniques

Abstract

Programming models for distributed and heterogeneous machines are rapidly growing in popularity to meet the demands of modern workloads. Task and actor models are common choices that offer different trade-offs between development productivity and achieved performance. Task-based models offer better productivity and composition of software, whereas actor-based models routinely deliver better peak performance due to lower overheads. While task-based and actor-based models appear to be different superficially, we demonstrate these programming models are duals of each other. Importantly, we show that this duality extends beyond functionality to performance, and elucidate techniques that let task-based systems deliver performance competitive with actor-based systems without compromising productivity. We apply these techniques to both Realm, an explicitly parallel task-based runtime, as well as Legion, an implicitly parallel task-based runtime. We show these techniques reduce Realm's overheads by between 1.7-5.3x, coming within a factor of two of the overheads imposed by heavily optimized actor-based systems like Charm++ and MPI. We further show that our techniques enable between 1.3-5.0x improved strong scaling of unmodified Legion applications.