Title :
Analyzing component-based systems using the self-organizing map
Author_Institution :
NOKIA Res. Center, Finland
fDate :
30 Aug.-3 Sept. 2005
Abstract :
As the complexity of component-based systems grows and their life cycles shorten, managing and understanding the dynamics and operation of the system at runtime becomes evermore difficult and at the same time more important. It is essential for system engineers to gain comprehensive understanding on not only the static but also the dynamic structure of the system. In this paper, we suggest a novel approach for studying the dynamic structure of a component-based system using the self-organizing map. We extract information about the communication traffic of the system into execution traces. These traces are then parsed and the communication traffic stored into a communication matrix. This matrix is used to construct a self-organizing map to describe the dynamic communication structure of the system.
Keywords :
object-oriented programming; self-organising feature maps; communication matrix; component-based system analysis; component-based system complexity; dynamic communication; dynamic component-based system; information extraction; parsing; self-organizing map; static component-based system; Application software; Context; Data mining; Information analysis; Simple object access protocol; Software engineering; Systems engineering and theory; TCPIP; Visualization;
Conference_Titel :
Software Engineering and Advanced Applications, 2005. 31st EUROMICRO Conference on
Print_ISBN :
0-7695-2431-1
DOI :
10.1109/EUROMICRO.2005.14