Towards an optimized software architecture for component adaptation at middleware level

Thomas Pramsohler, Simon Schenk, Uwe Baumgarten

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

4 Scopus citations

Abstract

The amount of software in the automotive domain is steadily increasing. Existing functions are adapted or enhanced on a regular basis. Often, such adaptations do not allow to keep the interfaces of the concerned components stable, leading to incompatibilities with former systems. In this contribution, we propose an optimized adaptation software architecture to deal with mismatching interfaces. We extend existing middleware solutions with transparent adapter loading capabilities. This enables for seamless adapter integration on those systems. As adapter model we use a finite-state machine aside with a domain specific language. By extracting static adaptations from the state machine we achieve state reduction and performance gain. The approach is evaluated using an automotive case-study.

Original languageEnglish
Title of host publicationSoftware Architecture - 7th European Conference, ECSA 2013, Proceedings
Pages266-281
Number of pages16
DOIs
StatePublished - 2013
Event7th European Conference on Software Architecture, ECSA 2013 - Montpellier, France
Duration: 1 Jul 20135 Jul 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7957 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference7th European Conference on Software Architecture, ECSA 2013
Country/TerritoryFrance
CityMontpellier
Period1/07/135/07/13

Keywords

  • adaptation architecture
  • behavioral adaptation
  • middleware adapter
  • software components
  • software composition

Fingerprint

Dive into the research topics of 'Towards an optimized software architecture for component adaptation at middleware level'. Together they form a unique fingerprint.

Cite this