This page is for discussions and ideas on our paper on Software Design as an Optimisation Problem from the Dagstuhl seminar 09432 "Quantitative Software Design". Feel free to add notes and ideas. You can register in our Wiki to avoid the captchas when editing and to create a correct history of this page. You can sign your comments using ~~~~ in the wiki source code, which becomes "martens 11:26, 7. Dez. 2009 (CET)", for example.

You can choose to observe this page and get notified by email by clicking the link "beobachten" (observe) above. You can edit this page by clicking "bearbeiten" (edit). More information: HowTo: Use A German Wiki

Discussion

Add any notes, comments, questions etc. here in a new subsection.

Call for Chapters and Information

Infos from Iman

Dear QSD Dagstuhl participant,

I write to you on behalf of the organizers of the Quantitative Software Design Dagstuhl seminar.

We are planning a book to be associated with the seminar, giving a state-of-the-art for the field. The draft call is at the end of this email.

Would you be interested in developing a paper a related theme for this book?

(I've written separate emails to individual participants, asking for their help in developing chapters specific to the breakout sessions.)

We cannot as yet promise a Springer publication, as we must go through a longer process now to get this approved. Our success will depend partly on having a high caliber author like yourself involved! On the other hand, the organizers have a good history with Springer.

Call for Chapters

QUANTITATIVE SOFTWARE DESIGN BOOK CALL FOR BOOK CHAPTERS

Between 20.10.09 and 23.10.09, Kresig, Reussner and Poernomo ran a seminar on Quantitative Software Design was held at the International Conference and Research Center (IBFI), Schloss Dagstuhl. The seminar's audience consisted of both academics and industrialists from software engineering and business informatics and led to some fascinating discussion and interdisciplinary interaction.

For this reason, the organizers felt it appropriate to organize an associated refereed book outlining the state-of-the-art on the topic, with invited contributions from both participants and experts who were unable to attend.

Quantitative software design

Quantitative software design is a field of research that is not yet firmly established. A number of challenging open research issues are only recently being addressed by the academic research community. The topic is also gaining increasing emphasis in industrial research, as any progress towards a more systematic and goal-driven software design promises the reduction of costs and risks of software projects, by avoiding current ³trial-and-error² approaches to design. The whole field is therefore of high industrial relevance, though it is far from providing ready-to-use solutions.

Topics of interest

Quantitative software design is a research area not yet firmly established. Its subject is the investigation of the relationship of the design of a software system on quantitatively measurable quality attributes. Such quality attributes include internal quality attributes (such as maintainability), but also externally measurable attributes (such as performance metrics, reliability or availability). This also includes quality attributes where quantitative metrics are under current investigation, such as security.

While there is no debate on the fact that the software design (mainly its architecture) is the main influencing factor on the quality of the resulting software system, an understanding of how an architecture¹s influence on the quality is currently primarily anecdotal.

Much progress was made on recent years in the area of model-based and model-driven quality prediction where software architectures are used as an input for the prediction of the quality of the system, namely various performance metrics, such as throughput, response time or reaction time. However, several important scientific questions remain unanswered:

trade-off decisions between antagonistic quality attributes
quantitative metrics for relevant quality attributes such as security
software design as an optimisation problem
lifting classical maintainability metrics to the architectural level

The aim of the book is to present the state-of-the-art from experts in relevant areas and consequently to define the research trajectory of the field of quantitative software design.

Relevant chapters are requested from experts in areas such as software architecture, component-based software engineering, model-based software quality prediction, machine learning, software performance, software reliability, software maintainability and security.

Dates

Chapter submission: 9th of April Review: 9th of May Publication: Near the end of 2010