Wenn Sie diesen Text sehen, ist auf ihrem Gerät noch nicht das neue Layout geladen worden. Bitte laden Sie diese Seite neu (ggf. mit gedrückter 'Shift'- oder 'Alt'-Taste) oder in einem 'privaten Fenster'.
Weitere Hinweise unter https://www.uni-hildesheim.de/wiki/lsf/faq/fehler.im.layout.

Zur Seitennavigation oder mit Tastenkombination für den accesskey-Taste und Taste 1 
Zum Seiteninhalt oder mit Tastenkombination für den accesskey und Taste 2 
Startseite    Anmelden     

A Commit-Based Analysis of Software Product Line Evolution: Two Case Studies (Forschungsbericht) - Einzelansicht


  • Funktionen:

Grunddaten

Titel der Arbeit (title) A Commit-Based Analysis of Software Product Line Evolution: Two Case Studies
Erscheinungsjahr 2017
Seitenzahl (pages) 1-57
Institution (institution) Software Systems Engineering, Institut für Informatik, Universität Hildesheim
Band-Heft-Nr 2/2017, SSE 2/17/E
Organisator (organization) University of Hildesheim
Publikationsart Forschungsbericht
Inhalt
Abstract Software Product Line (SPL) evolution affects a variety of artifact types, each containing their artifact-specific information as well as variability information. While the former information type defines the basic content of an artifact, like the general program definition in code artifacts, the latter supports the customization of these artifacts for different products of the SPL. Existing work that aims at characterizing the state and evolution of a product line identifies general SPL evolution scenarios or evaluates the (co-)evolution of variability information in different types of artifacts. However, these results are typically based on a feature perspective, which abstracts from the SPL evolution at large. We argue that an artifact-based analysis of product line evolution will complement existing work by analyzing the intensity of changes to different artifact and information types. In this report, we therefore present an approach for and the results of the extraction and analysis of changes introduced to artifact-specific and variability information in code, build and variability model artifacts. This approach has been developed for and applied to the Linux kernel. In order to broaden our analysis, we also apply the approach to the Coreboot firmware. The results reveal the intensity of changes over the evolution history of both SPLs. Further, we compare the intensity of changes across both SPLs with respect to the size of the changes as well as over time.

Beteiligte Personen

Kröher, Christian  M.Sc.  
Schmid, Klaus, Professor Dr. rer. nat.  

Einrichtung

Abt. Software Systems Engineering

Schlüsselwörter

Projekt REVaMP²
Impressum      Datenschutzerklärung     Datenschutz      Datenschutzerklärung     Erklärung zur Barrierefreiheit