Ray: A Distributed Framework For Emerging AI Applications

Authors:
Philipp Moritz UC Berkeley
Robert Nishihara UC Berkeley
Michael I. Jordan UC Berkeley
Robert Nishihara UC Berkeley
Stephanie Wang UC Berkeley
Alexey Tumanov UC Berkeley
Richard Liaw UC Berkeley
Eric Liang UC Berkeley
Melih Elibol UC Berkeley
Zongheng Yang UC Berkeley
William Paul UC Berkeley

Introduction:

The next generation of AI applications will continuously interact with the environment and learn from these interactions.These applications impose demanding systems requirements.In this paper, the authors consider these requirements and present Ray — a distributed system to address them. Ray implements a unified interface that can express both task-parallel and actor-based computations, supported by a single dynamic execution engine.

Abstract:

The next generation of AI applications will continuously interact with the environment and learn from these interactions. These applications impose new and demanding systems requirements, both in terms of performance and flexibility. In this paper, we consider these requirements and present Ray — a distributed system to address them. Ray implements a unified interface that can express both task-parallel and actor-based computations, supported by a single dynamic execution engine. To meet the performance requirements, Ray employs a distributed scheduler and a distributed and fault-tolerant store to manage the system’s control state. In our experiments, we demonstrate scaling beyond 1.8 million tasks per second and better performance than existing specialized systems for several challenging reinforcement learning applications.

You may want to know: