| 摘要 |
PROBLEM TO BE SOLVED: To improve the reliability of a cluster by sending a previously generated relief message to other nodes of a cluster when a fault phenomenon is detected. SOLUTION: A main memory 120 properly includes one or multiple application programs 121, a cluster management application 122, and an operating system 123 including a cluster-node relief signal generating means 124. This cluster-node relief signal generating means 123 provides a mechanism which needs to send a relief signal to other nodes in a cluster in the case of a serious fault occurring to a node. Namely, the cluster-node relief signal generating means 124 integrated with the operating system 123 includes a relief signal sending-out method, a previously generated relief message, and a dedicated relief signal for sending the message. Then when the fault phenomenon is detected, the previously generated relief message is sent to other nodes of the cluster.
|
