A unifying approach to data-independence

34Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.
Get full text

Abstract

A concurrent system is data-independent with respect to a data type when the only operation it can perform on values of that type is equality testing. The system can also assign, input, nondeterministi-cally choose, and output such values. Based on this intuitive definition, syntactic restrictions which ensure data-independence have been formulated for a variety of different formalisms. However, it is difficult to see how these are related. We present the first semantic definition of data-independence which allows equality testing, and its extension which allows constant symbols and predicate symbols. Both are special cases of a definition of when a family of labelled transition systems is parametric. This provides a unified approach to data-independence and its extensions. The paper also contains two theorems which, given a system and a specification which are data-independent, enable the verification for all instantiations of the data types (and of the constant symbols and the predicate symbols, in the case of the extension) to be reduced to the verification for a finite number of finite instantiations. We illustrate the applicability of the approach to particular formalisms by a programming language similar to UNITY. © Springer-Verlag Berlin Heidelberg 2000.

Cite

CITATION STYLE

APA

Lazic, R., & Nowak, D. (2000). A unifying approach to data-independence. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1877 LNCS, pp. 581–596). Springer Verlag. https://doi.org/10.1007/3-540-44618-4_41

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free