An approach to modelling and analysing reliability of Breeze/ADL-based software architecture

Chen Li, Hong Ji Yang*, Hua Xiao Liu

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

1 Citation (Scopus)

Abstract

Breeze/architecture description language (ADL), is an eXtensible markup language (XML) based architecture description language which is used to model software systems at the architecture level. Though Breeze/ADL provides an appropriate basis for architecture modelling, it can neither analyse nor evaluate the architecture reliability. In this paper, we propose a Breeze/ADL based strategy which, by combining generalized stochastic Petri net (GSPN) and tools for reliability analysis, supports architecture reliability modelling and evaluation. This work expands the idea in three directions: Firstly, we give a Breeze/ADL reliability model in which we add error attributes to Breeze/ADL error model for capturing architecture error information, and at the same time perform the system error state transition through the Breeze/ADL production. Secondly, we present how to map a Breeze/ADL reliability model to a GSPN model, which in turn can be used for reliability analysis. The other task is to develop a Breeze/ADL reliability analysis modelling tool–EXGSPN (Breeze/ADL reliability analysis modelling tool), and combine it with platform independent petri net editor 2 (PIPE2) to carry out a reliability assessment.

Original languageEnglish
JournalInternational Journal of Automation and Computing
Volume14
Issue number3
Pages (from-to)275-284
Number of pages10
ISSN1476-8186
DOIs
Publication statusPublished - 1 Jun 2017
Externally publishedYes

Keywords

  • Breeze graph grammar
  • Breeze/architecture description language (ADL)
  • generalized stochastic Petri net (GSPN)
  • reliability
  • Software architecture

Fingerprint

Dive into the research topics of 'An approach to modelling and analysing reliability of Breeze/ADL-based software architecture'. Together they form a unique fingerprint.

Cite this