DocumentCode :
3044338
Title :
Performance prediction based on semi-formal software architectural description
Author :
Li, J. Jenny
Author_Institution :
Bellcore, Morristown, NJ, USA
fYear :
1998
fDate :
16-18 Feb 1998
Firstpage :
193
Lastpage :
199
Abstract :
Mature engineering includes systematic quantified architectural design. Software engineering lacks engineering methodologies in architectural design and the evaluation of the design in terms of performance. This paper presents a methodology and its accompanying toolset, Workflow-to-SDL Direct Simulation (W2S), for generating a simulator from a semi-formal architecture description. We have used this simulator to study the behavior and reliability of the system. This paper focuses on the performance aspect of the system. We used the simulator to predict the system´s performance before the system is built to guild us in the selection of architecture designs. The focus of our research is the telecommunication domain. The W2S methodology was applied to a network resource management system, using a set of tools that we developed. Several sets of experiments were carried out. The result shows that the information generated by the simulator can be used to predict the performance of a system solely based on its architectural description. It allows designers to select architectures with better performance
Keywords :
computer network management; finite state machines; formal specification; software performance evaluation; telecommunication computing; virtual machines; W2S methodology; Workflow-to-SDL Direct Simulation; architecture selection; communicating extended finite state machine; mature engineering; network resource management system; semi-formal software architectural description; simulator generation; software architecture; software engineering; software performance prediction; system behavior; system reliability; systematic quantified architectural design; telecommunication; Computer architecture; Design engineering; Design methodology; Predictive models; Reliability engineering; Resource management; Software engineering; Software performance; System performance; Systems engineering and theory;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Performance, Computing and Communications, 1998. IPCCC '98., IEEE International
Conference_Location :
Tempe/Phoenix, AZ
ISSN :
1097-2641
Print_ISBN :
0-7803-4468-5
Type :
conf
DOI :
10.1109/PCCC.1998.659951
Filename :
659951
Link To Document :
بازگشت