The Free Termination Property of Queries Over Time

TL;DR

This paper investigates conditions under which distributed database nodes can unilaterally terminate queries without coordination, introducing a new semiautomata-based model linking relational transducers and algebraic models like CRDTs.

Contribution

It defines the free termination property for distributed queries and develops a semiautomata model connecting CALM transducers with algebraic frameworks.

Findings

Characterization of free termination conditions

Development of a semiautomata-based model

Bridging relational and algebraic models

Abstract

Building on prior work on distributed databases and the CALM Theorem, we define and study the question of free termination: in the absence of distributed coordination, what query properties allow nodes in a distributed (database) system to unilaterally terminate execution even though they may receive additional data or messages in the future? This completeness question is complementary to the soundness questions studied in the CALM literature. We also develop a new model based on semiautomata that allows us to bridge from the relational transducer model of the CALM papers to algebraic models that are popular among software engineers (e.g. CRDTs) and of increasing interest to database theory for datalog extensions and incremental view maintenance.