Based On Reflected Publish/Subscribe System Status Monitoring Technology Research

Publish/subscribe system belongs to distributed computing range, the system has acertain degree of complexity. Monitoring publish/subscribe transceiver status and connectionrelationship of the various participants in the system is a very important part of, it can beconducive to system administrators to properly manage the system and observe the operatingstatus of each participant. Publish/subscribe system monitoring technology for the study ofgreat significance.This article first carry out a detailed study of existing monitoring technology in datacollection methods, combined with the characteristics of publish/subscribe systems,proposed the base publish/subscribe data collection methods.Second, because this method need use reflection principle to complete data update,therefore, this paper studies the dynamic configuration and reflection principle, propose thereflection-based monitoring system dynamic configuration model, the model is divided intodynamic configuration and other functional parts, combined with the reflection principle tobuild a model of the reflection of the monitor model. And in the monitoring model eachcomponent model are described in detail.Again, this paper proposes a specific topic-based trigger for publish/subscribe systemsdata update algorithm, in order to solve the monitoring data update problem, on this basis thepropose Delete topics algorithm, create topic algorithm, delete the monitoring componentalgorithm and create a monitoring component algorithm four algorithms. After two algorithmsneed to use the stationary state theory to ensure the consistency of the system before andafter the dynamic configuration.Finally, based on APPDDS platform verify the correctness of the monitor model andalgorithm, by on in APPDDS platforms to achieve the monitoring system based on the model,and successfully obtain the required monitoring data to apply the appropriate algorithm tocomplete the operation of the dynamic update of data and when the change in the participant.