Exception handling is an important issue in dependable software computing. This paper presents a system architecture and explains the role of responsibility of each component that is introduced in the architecture. The design is used to ensure the constructed service-oriented applications remain in a stable state. Service-Oriented Architecture is a paradigm for constructing current enterprise applications. It uses service composition technology to induce existing service components to provide advanced services. The most important concept in this methodology is that of reusing existing components. Business Process Execution Language (BPEL) is widely used for achieving service composition, providing basic capability for developers to deal with faults using defined activities standard in BPEL. No sufficient design, however, supports developers in implementing fault-handling components efficiently. The architecture proposed in this paper provides the design approach to explain how to reuse the fault-handling components. Additionally, the paper also provides an empirical case study that illustrates how this design was used by the Chunghwa Telecom Company in an effort to improve efficiency in implementing fault-handling components and in constructing service-oriented applications. The number of fault-handling components required was reduced from 38 to 3, implying a 92.10% efficiency rate.

1. [1] G. Laures, “Are Service-Oriented Architectures the Panacea for a High-Availability Challenge,” Lecture Notes in Computer Science, LNCS 3694, pp. 102-106, 2005.
2. [2] N. Milanovic and M. Malek, “Service-Oriented Operating System-A Key Element in Improving Service Availability,” Lecture Notes in Computer Science, LNCS 4526, pp. 31-42, 2007.
3. [3] N. Salatge and J. Fabre, “Fault Tolerance Connectors for Unreliable Web Services,” in IEEE Int. Conf. on Dependable Systems and Networks, pp. 51-60, 2007.
4. [4] D. Jayasinghe, “FAWS for SOAP-based Web Services : A clienttransparent fault tolerance system for SOAP-based Web Services,” IBM developer Works, 2005.
   http://www.ibm.com/developerworks/webservices/library/ws-faws/
5. [5] W. Zhao, H. Zhang, and H. Chai, “A Lightweight Fault Tolerance Framework for Web Services,” Web Intelligence and Agent Systems, pp. 255-268, 2009.
6. [6] W. Zhao, “BFT-WS: A Byzantine Fault Tolerance Framework for Web Services,” in IEEE EDOC Conf. The Enterprise Computing Conf., pp. 89-96, 2007.
7. [7] W. Tan, L. Fong, and N. Bobroff, “BPEL4Job: A Fault-Handling Design for Job Flow Management,” Leture Notes in Computer Science, LNCS 4749, pp. 27-42, 2007.
8. [8] J. L. Knudsen, “Exception handling versus fault tolerance,” in Exception Handling, in Object Oriented Systems Workshop at ECOOP’2000, pp. 16-31, 2000.
9. [9] A. Liu, Q. Li, L. Huang, and M. Xiao, “A Declarative Approach to Enhancing the Reliability of BPEL Process,” in IEEE Int. Conf. on Web Services, pp. 272-279, 2007.
10. [10] Y. I. Chen, G. K. Ni, and C. Y. Lin, “A runtime-adaptable service bus design for telecom operations support systems,” IBM Systems J., Vol.47, No.3, pp. 445-456, August 2008.