Analyzing the use of various software system development techniques to show productivity gains or losses ( i.e. , typically concerned with the productivity of the software developers).
Published in Chapter:
Quantitative Productivity Analysis of a Domain-Specific Modeling Language
Joe Hoffert (Indiana Wesleyan University, USA), Douglas C. Schmidt (Vanderbilt University, USA), and Aniruddha Gokhale (Vanderbilt University, USA)
Copyright: © 2015
|Pages: 32
DOI: 10.4018/978-1-4666-6359-6.ch013
Abstract
Model-Driven Engineering (MDE), in general, and Domain-Specific Modeling Languages (DSMLs), in particular, are increasingly used to manage the complexity of developing applications in various domains. Although many DSML benefits are qualitative (e.g., ease of use, familiarity of domain concepts), there is a need to quantitatively demonstrate the benefits of DSMLs (e.g., quantify when DSMLs provide savings in development time) to simplify comparison and evaluation. This chapter describes how the authors conducted quantitative productivity analysis for a DSML (i.e., the Distributed Quality-of-Service [QoS] Modeling Language [DQML]). The analysis shows (1) the significant quantitative productivity gain achieved when using a DSML to develop configuration models compared with not using a DSML, (2) the significant quantitative productivity gain achieved when using a DSML interpreter to automatically generate implementation artifacts as compared to alternative methods when configuring application entities, and (3) the viability of quantitative productivity metrics for DSMLs.