Partially-Distributed Coordination with Reo

Abstract

Coordination languages, as Reo, have emerged for the specification and implementation of interaction protocols among concurrent entities. In this paper, we propose a framework for generating partially-distributed, partially-centralized implementations of Reo connectors to improve 1) build-time compilation and 2) run-time throughput and parallelism. Our framework relies on the definition of a new formal product operator on constraint automata (Reo's formal semantics), which enables the formally correct distribution of disjoint parts of a coordination scheme over different machines according to several possible motivations (e.g., performance, privacy, QoS constraints, resource availability, network topology). First, we describe the design and a proof-of-concept implementation of our framework. Then, in a case study, we show and explain how a generated connector implementation can be executed in the Cloud and supports Big Data coordination.