Monitoring Design And Implementation Of A Central Server Within The Network Monitoring System

With the rapid development of computer technology, as well as the growing size of the network, the network has spread into various industries and people's lives. Today, information society becomes inseparable from the network. And the stability of network and reliability of its operation has become important factors that can not be neglected. In order to improve network management, network monitoring can be a necessary means. With the help of network monitoring system, we can acquire a variety of performance data and analyze the operating conditions, so the problems could be detected timely. This provides network administrators with a convenient and efficient way of management.We mainly study the center server of the inner-network monitoring system in this thesis. As the project background, the inner-network monitoring system integrates the merits of centralized monitoring and distributed monitoring. And it focuses on the idea of centralized management and distributed monitoring. So network monitoring can be achieved by deploying multiple monitoring agents. Since the monitoring center server takes the role of control center, it is an important part of the whole system. In the process of realization, we divide the structure of server into six modules based on the principle of modular design. And the first three modules reflect the core function of server. Generally speaking, we mainly do the following work in the thesis:1. We analyze some key technologies in implementing the monitoring center server, and then design the structure of monitoring center server and its function modules.2. According to the communication features, we work out a set of communications rules by adopting TCP protocol and a regular head of message. It is used to achieve interaction between the server and agent.3. For the concurrency issues in communication, we design a special method. The feature of it is that we could handle collected data from multiple agents by introducing multi-threads at the same time. So it can increase overall system capacity of parallel processing.4. A storing approach with a buffer is designed for the purpose of optimizing the efficiency in storing data. Firstly, all collected data are received into a buffer by worker threads. Then, a special thread would regularly get data from the buffer to achieve the storage.Finally, we test the server in many aspects. And the results show that the system can effectively run under the premise of realizing function.