Extending Fault Tolerance Patterns by Visual Degradation Rules (bibtex)
by ,
Abstract:
Embedded distributed systems play an important role in many advanced technical systems. In order to satisfy high availability and reliability requirements, fault tolerance techniques such as triple modular redundancy are employed. In addition, techniques for graceful degradation are required to handle situations when a system experiences too many faults to compensate them while still providing a reduced albeit sufficient functionality. As a formal visual specification technique to describe known standard fault tolerance solutions we proposed fault tolerance patterns [24] which capture the essential structure and relevant deployment restrictions of these solutions. Fault tolerance patterns are easily applied during the design of component-based systems to increase the reliability or availability of specific components or subsystems and permit to derive a correct initial deployment and guide the self-repair of the system. In this paper, we extend our fault tolerance pattern approach with additional visual degradation rules. The rules can at first be employed to define reconfiguration steps for the system which reduce the provided level of fault tolerance while retaining the provision of functional properties. Secondly, steps which result in a graceful degradation and thus only a reduced functionality can be defined.
Reference:
Extending Fault Tolerance Patterns by Visual Degradation Rules (Matthias Tichy, Holger Giese), In Proc. of the Workshop on Visual Modeling for Software Intensive Systems (VMSIS) at the the IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC'05), Dallas, Texas, USA, 2005.
Bibtex Entry:
@InProceedings{TG05_ag,
AUTHOR = {Tichy, Matthias and Giese, Holger},
TITLE = {{Extending Fault Tolerance Patterns by Visual Degradation Rules}},
YEAR = {2005},
MONTH = {September},
BOOKTITLE = {Proc. of the Workshop on Visual Modeling for Software Intensive Systems (VMSIS) at the the IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC'05), Dallas, Texas, USA},
PAGES = {67-74},
URL = {http://www.upb.de/cs/ag-schaefer/Veroeffentlichungen/Quellen/Papers/2005/TG05.pdf},
PDF = {TG05.pdf},
ABSTRACT = {Embedded distributed systems play an important role in many advanced technical systems. In order to satisfy high availability and reliability requirements, fault tolerance techniques such as triple modular redundancy are employed. In addition, techniques for graceful degradation are required to handle situations when a system experiences too many faults to compensate them while still providing a reduced albeit sufficient functionality. As a formal visual specification technique to describe known standard fault tolerance solutions we proposed fault tolerance patterns [24] which capture the essential structure and relevant deployment restrictions of these solutions. Fault tolerance patterns are easily applied during the design of component-based systems to increase the reliability or availability of specific components or subsystems and permit to derive a correct initial deployment and guide the self-repair of the system. In this paper, we extend our fault tolerance pattern approach with additional visual degradation rules. The rules can at first be employed to define reconfiguration steps for the system which reduce the provided level of fault tolerance while retaining the provision of functional properties. Secondly, steps which result in a graceful degradation and thus only a reduced functionality can be defined.}
}
Powered by bibtexbrowser