Design Of Server Monitoring System Based On Cloud Computing Platform

Nowadays, with the rapid development of cloud computing technology, cloud data centers are increasing. Cloud data center has lots of servers. Monitoring the server’s status has become a very important part of the cloud platform. A good monitoring system can ensure the stability of cloud platform and increase the availability of cloud services. However, considering the huge number of servers in the cloud data center, traditional monitoring systems are difficult to ensure a real-time and efficiency monitoring. So, it is meaningful to study on a real-time, efficiency and low occupancy server monitoring system.Cloud computing technology provides a new platform for the development of server monitoring technology. And JMX has been used in the structure of the monitoring systems. We combined JMX and cloud computing technology and studied a server monitoring system for cloud data center which had a large number of servers.Firstly, we described the architecture of server monitoring system. Considering the number of servers in the cloud data center is huge, we have adopted a hierarchical structure in our system. And we divided the monitoring system into monitoring server, data collectors and monitoring agents.Secondly, the article has studied data collection model and data processing model of the monitoring system. It introduced their research status and proposed the corresponding solutions. For data collection model, we designed a hybrid transmission mode based on failure rate threshold. And we implemented the model of data processing based on the Map/Reduce programming model.Finally, according to the distributed storage features of the cloud computing platform, we optimized the structure of data storage to reduce the complexity of distributed data processing. We made an assessment of the monitoring systems, and the assessment indicated that the system tooks up less server resources and had a high real-time.