Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15] [ 16] [ 17] [ 18] [ 19]


Journal of Information Science and Engineering, Vol. 25 No. 6, pp. 1753-1781 (November 2009)

An Automatic Code Generation for Self-Healing*

?School of Information and Communication Engineering
Sungkyunkwan University
Suwon 400-746, South Korea

The overall goal of this research is to improve self-healing development environments to facilitate self-managing capabilities such as self-healing. Self-healing is an approach to evaluating constraints defined in a target system and to apply an appropriate strategy when the constraints have been violated. Much attention has recently been focused on self-healing ability that recognizes problems arising in a target system. However, in order to generate code for self-healing, most of the existing approaches show that self-healing development environments need much effort and time to analyze and model constraints. Hence, for self-healing, this paper proposes an automated code generation approach to determine problems arising in external and internal system environments. The approach proposes: (1) Specifying the target system through the models created in the design phase of target system. (2) Automatically creating constraints for the external and internal system environment by using the specified contents. (3) Deriving a dependency model of a component based on the created internal state rule. (4) Translating the constraints and dependency model into code evaluating behaviors of the target system, and the determining problem level. (5) Monitoring an internal and external status of system based on the level of problem determination, and applying a self-healing strategy when detecting an abnormal state caused in the target system. As proof of the approach, we used a video conference system and E-Commerce System for the case study. The experimental results show that it achieves reducing the efforts required for developers of self-healing systems to analyze the target system and resolving abnormal behavior of the target system in regard to both external and internal problem.

Keywords: code generation, autonomic computing, self-healing, external state, internal state, error detection

Full Text () Retrieve PDF document (200911_07.pdf)

Received January 9, 2008; revised August 12 & October 8 & November 21, 2008; accepted December 11, 2008.
Communicated by Jonathan Lee.
* This paper was supported by Faculty Research Fund, Sungkyunkwan University, 2008.