Deriving work plans for solving performance and scalability problems (bibtex)
by Heger, Christoph and Heinrich, Robert
Abstract:
The performance of an enterprise application (e.g. response time, throughput, or resource utilization) is an important quality attribute that can have a significant impact on a company's success. When a performance problem such as a performance bottleneck has been detected, the root cause identified and a solution proposed, developers have to identify the elements of the application often manually that will undergo changes and determine how these elements must be changed in order to implement the solution. Many existing approaches are able to identify the elements that have to be modified but only few are able to determine the necessary types of changes on these elements. Neither of the approaches supports developers with a work plan sketching the implementation steps. In this paper, we propose an approach to point developers the way torwards an implementation of a performance or scalability solution with an ordered set of work activities. Rules are used to derive a work plan sketching the implementation of a solution for the particular application based on an initial set of work activities. The rule-based approach identifies impacted elements and determines how they should be changed. We demonstrate the proposed approach with a solution of a performance bottleneck as an example. \textcopyright 2014 Springer International Publishing.
Reference:
Deriving work plans for solving performance and scalability problems (Heger, Christoph and Heinrich, Robert), Chapter in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer International Publishing, volume 8721 LNCS, 2014.
Bibtex Entry:
@incollection{heger2014deriving,
abstract = {The performance of an enterprise application (e.g. response time, throughput, or resource utilization) is an important quality attribute that can have a significant impact on a company's success. When a performance problem such as a performance bottleneck has been detected, the root cause identified and a solution proposed, developers have to identify the elements of the application often manually that will undergo changes and determine how these elements must be changed in order to implement the solution. Many existing approaches are able to identify the elements that have to be modified but only few are able to determine the necessary types of changes on these elements. Neither of the approaches supports developers with a work plan sketching the implementation steps. In this paper, we propose an approach to point developers the way torwards an implementation of a performance or scalability solution with an ordered set of work activities. Rules are used to derive a work plan sketching the implementation of a solution for the particular application based on an initial set of work activities. The rule-based approach identifies impacted elements and determines how they should be changed. We demonstrate the proposed approach with a solution of a performance bottleneck as an example. {\textcopyright} 2014 Springer International Publishing.},
author = {Heger, Christoph and Heinrich, Robert},
booktitle = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
doi = {10.1007/978-3-319-10885-8_8},
isbn = {9783319108841},
issn = {16113349},
keywords = {Impact Propagation,Rules,Software Performance Engineering,Solution Implementation Support,iobserve},
mendeley-tags = {iobserve},
pages = {104--118},
publisher = {Springer International Publishing},
title = {{Deriving work plans for solving performance and scalability problems}},
volume = {8721 LNCS},
year = {2014}
}
Powered by bibtexbrowser