DiscoTect: A System for Discovering Architectures from Running Systems

  • Authors:
  • Hong Yan;David Garlan;Bradley Schmerl;Jonathan Aldrich;Rick Kazman

  • Affiliations:
  • Carnegie Mellon University;Carnegie Mellon University;Carnegie Mellon University;Carnegie Mellon University;Carnegie Mellon University

  • Venue:
  • Proceedings of the 26th International Conference on Software Engineering
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

One of the challenging problems for software developersis guaranteeing that a system as built is consistentwith its architectural design. In this paper we describe atechnique that uses run time observations about an executingsystem to construct an architectural view of thesystem. With this technique we develop mappings thatexploit regularities in system implementation and architecturalstyle. These mappings describe how low-levelsystem events can be interpreted as more abstract architecturaloperations. We describe the current implementationof a tool that uses these mappings, and show that itcan highlight inconsistencies between implementationand architecture.