Mastersthesis,
Extending the Descartes Query Language to Support Automated What-If Analysis
University of Würzburg, Am Hubland, Informatikgebäude, 97074 Würzburg, Germany, Bachelor Thesis, (August 2016)
Abstract
For modern systems, performance engineering is important. Performance Engineering is quite difficult because there are many different analysis aproaches. The Descartes Query Language has been developed to support an easy usage of performance analysis approaches. It enables the user to separate the definition of the users concern and its deduction. In this thesis, DQL gets an extension to support "What if...?" analyses. WhatIFQueries are designed to be as simple and short as possible as they require less knowledge about the system/model of the user than existing query types. Still, they allow more complex structures to get different result outputs. This thesis contributes to query for all parameter changes without the need of specifying them. It also adds filtering mechanisms and functionality to automatically compare internal and external variations to current state. The evaluation for the implementation was done with connectors for several solution approaches and provides unit tests.
