云环境下节点失效检测机制研究  

Research of Node Failure Detection Mechanism in Cloud Environment

作　　者：刘艾侠[1] 

机构地区：[1]宝鸡职业技术学院计算机系,宝鸡721013

出　　处：《微型电脑应用》2017年第5期59-62,共4页Microcomputer Applications

摘　　要：针对云计算系统的可靠性问题,设计云环境下基于拜占庭容错模型的节点失效检测机制。当云环境系统中发生节点失效或者节点超时,利用设计的拜占庭容错算法,自动检测出系统的失效节点并且隔离,提高系统的检测精度。算法通过构造节点矩阵检测出失效节点。仿真实验表明,该模型能够有效的隔离云环境中失效节点和超时节点,大大提高了系统的可靠性。该模型还提供了反向恢复,当所有节点通过拜占庭容错模块,若失效节点数大于总节点数的1/3,系统将反向恢复到输入缓冲器。In view of the problem of reliability of cloud computing system, node failure detection mechanism based on Byzantine fault-tolerant model in cloud environment is designed. When the node failure or time-out occurs in cloud environment system, the designed Byzantine fault-tolerant algorithm is utilized to automatically detect and isolate system failure node, and improve the test accuracy. The failure node is detected by constructing node matrix. Simulation experiments show that the model can effectively isolate node failure or timeout in cloud environment, greatly improving the reliability of the system. The model also provides a reverse recovery. When all the nodes go through the Byzantine fault-tolerant module, if the failure node number is more than one third of the total number of nodes, the system will reverse back to the input buffer.

关 键 词：云计算 失效检测 容错算法 缓冲器 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]