Discovering pattern-based mediator services from communication logs

C. Gierds, D. Fahland

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Process discovery is a technique for deriving a conceptual high-level process model from the execution logs of a running implementation. The technique is particularly useful when no high-level model is available or in case of significant gaps between process documentation and implementation. The discovered model makes the implementation accessible to various kinds of analysis for functional and non-functional properties. In this paper we extend process discovery to mediator services (or adapters) which adapt the messaging protocols of 2 or more otherwise incompatible services. We propose a technique that takes as input logs of communication behaviors—one log for each service connected to the adapter—and a library of high-level data transformation rules relevant for the domain of the adapter, and then returns an operational adapter model describing the control-flow and the data flow of the adapter in terms of Coloured Petri Nets – if such model exists. We discuss benefits and limitations of this idea and evaluate it with a prototype implementation on industrial size models. Keywords: Process Mining; Service Mining; Pattern Based Design; Coloured Petri nets; synthesis
Original languageEnglish
Title of host publicationService-Oriented Computing – ICSOC 2013 Workshops (CCSA, CSB, PASCEB, SWESE, WESOA, and PhD Symposium, Berlin, Germany, December 2-5, 2013. Revised Selected Papers)
EditorsA.R. Lomuscio, S. Nepal, F. Patrizi, B. Benatallah, I. Brandic
Place of PublicationBerlin
PublisherSpringer
Pages123-134
ISBN (Print)978-3-319-06858-9
DOIs
Publication statusPublished - 2014

Publication series

NameLecture Notes in Computer Science
Volume8377
ISSN (Print)0302-9743

Fingerprint

Dive into the research topics of 'Discovering pattern-based mediator services from communication logs'. Together they form a unique fingerprint.

Cite this