Network algebra for synchronous dataflow

TL;DR

This paper develops an algebraic framework for modeling synchronous dataflow networks, introducing Basic Network Algebra and extending it with branching constants, supported by models based on stream transformers and process algebra.

Contribution

It introduces a formal algebraic theory for synchronous dataflow networks, including new constants and axioms, with two models for the theory.

Findings

Algebraic properties of networks are captured by BNA.

Extended theory includes branching connection constants.

Models based on stream transformers and process algebra are provided.

Abstract

We develop an algebraic theory of synchronous dataflow networks. First, a basic algebraic theory of networks, called BNA (Basic Network Algebra), is introduced. This theory captures the basic algebraic properties of networks. For synchronous dataflow networks, it is subsequently extended with additional constants for the branching connections that occur between the cells of synchronous dataflow networks and axioms for these additional constants. We also give two models of the resulting theory, the one based on stream transformers and the other based on processes as considered in process algebra.