¹Shengda Economics Trade & Management College, Zhengzhou University, Zhengzhou, China 451191
²School of Information Engineering, Zhengzhou University, Zhengzhou 450001, China
³School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China
⁴Henan Provincial Key Lab on Information Network, Zhengzhou 450052, China

---

Abstract—Failure detection is a key technology in tolerant system. Failure detectors without adaptive mechanism cannot meet the requirements of QOS (quality of service) of applications because of the variations of the network in actual distributed system. Adaptive failure detectors should dynamically adjust the detecting quality according to the variations of the real-time state of the network. Assuming that the delay and loss of the messages is a random probability, a failure detection model based on the predicted message delay is proposed in this paper. A PAC-AFD adaptive failure detection algorithm is realized based on the above model which is on the basis of the prediction from historical message delay and contains checking idea. Experimental results show that the algorithm can relieve the effect of the delay and loss of the message on the failure detection while ensuring the accuracy and completeness of detection.

Index Terms—failure detection, QOS, distributed system, adaptive, checking.

[PDF]

Cite: Bin Liu, Shifei Yang, Lei Shi, Xiaoguang Ding, Qian Zhang, "Modeling of Failure Detector Based on Message Delay Prediction Mechanism," Journal of Software vol. 6, no. 9, pp. 1821-1828, 2011.