2004IEEE International Software Metrics SymposiumRequires access

Error propagation in software architectures

Walid Abdelmoez, D.M. Nassar, M. Shereshevsky, N. Gradetsky, R. Gunnalan, H.H. Ammar, Bo Yu, A. Mili

Open publisher page 72 citations

Abstract

The study of software architectures is emerging as an important discipline in software engineering, due to its emphasis on large scale composition of software products, and its support for emerging software engineering paradigms such as product line engineering, component based software engineering, and software evolution. Architectural attributes differ from code-level software attributes in that they focus on the level of components and connectors, and that they are meaningful for an architecture. In this paper, we focus on a specific architectural attribute, which is the error propagation probability throughout the architecture, i.e. the probability that an error that arises in one component propagates to other components. We introduce, analyze, and validate formulas for estimating these probabilities using architectural level information.

About this research paper

What this paper is about

The study of software architectures is emerging as an important discipline in software engineering, due to its emphasis on large scale composition of software products, and its support for emerging software engineering paradigms such as product line engineering, component based software engineering, and software evolution. Architectural attributes differ from code-level software attributes in that they focus on the level of components and connectors, and that they are meaningful for an architecture. In this paper, we focus on a specific architectural attribute, which is the error propagation probability throughout the architecture, i.e. the probability that an error that arises in one component propagates to other components. We introduce, analyze, and validate formulas for estimating these probabilities using architectural level information.

Why it matters

OpenAlex reports 72 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The study of software architectures is emerging as an important discipline in software engineering, due to its emphasis on large scale composition of software products, and its support for emerging software engineering paradigms such as product line engineering, component based software engineering, and software evolution. Architectural attributes differ from code-level software attributes in that they focus on the level of components and connectors, and that they are meaningful for an architecture. In this paper, we focus on a specific architectural attribute, which is the error propagation probability throughout the architecture, i.e. the probability that an error that arises in one component propagates to other components. We introduce, analyze, and validate formulas for estimating these probabilities using architectural level information.

Key concepts: Computer science, Software sizing, Software engineering, Software product line, Software construction, Software architecture, Resource-oriented architecture, Focus (optics)

Related papers

Back to paper searchBrowse research topicsOriginal source
Error propagation in software architectures — Research Paper | ScholarLens