A Design And Implementation Of Diagnostic System In Distributed Environment

With the rapid development of hardware and software technology and internet penetration for all walks of life,different company's products are deployed in diverse geographies and various work environment of different platform.The rapid expansion of the size of products brings a more complex communication environment and huge associated data.These data are very useful for analyzing the state of products,locating the exception of products,finding the disadvantages of products.A targeted tool can help users,developers,testers to find information more quickly and provide them some guiding solutions,thus saving a lot of manpower,resources and time.The diagnostics system described in this paper is such a tool.It has a good prospect in the age of big data.It can solve the contradiction between a large variety of data and limited human resources.In this paper,the author analyzes the company's current product,designs the structure of the diagnostic system from aspects of data share,resource exchange,information filtering and so on,finally get a diagnostic system that can work well in both Windows and Linux environment.Focusing on functions of data collection,data analysis and data show,this paper tells the design and implementation of data source proxy server,system server and client in the diagnostic system.The implementation of the system is showed with c++.In order to ensure the portability between Windows and Linux system,the similar functions in Windows and Linux are summarized during the implementation of diagnostic system.In addition,the author follows the idea of object-oriented programming and makes rational classes in the code.The author's major contributions are outlined as follows:1.In the process of the data source proxy server's realization.The proxy server is designed according to the knowledge of NT Service and daemon to ensure its function of data collection has nothing to do with the current user of the system and the input from the console of computer.Use tools such as wireshark to collect relevant data of the network with the knowledge of multi-process programming.Control the data collection work of log4cxx by modifying the relevant files.Create a TCP server to send response to the control information from the system server.2.The implementation of system server is based on the knowledge of TCP server,multi-thread programming and socket programming,the server can process messages get from the client.Define and implement a communication protocol to represent for the information transmission work between the system server and the data source proxy server.Use libcurl to make data share between system server and data source proxy server and transfer data that is processed with the defined rules to the client.3.In the implementation of client.Firstly,read configuration information from the relevant files.Secondly,use a protocol that is similar to cobra to complete the communication between the client and system server.Finally,output control information to the front-end that is made by python.Generic programming and design patterns are used to ensure the portability,scalability,reusability of the code during the implementation of the system.The system is tested and works well.