| With the continuous development of computer technology, the daily computerized requirements of the commercial and social institutions reached an unprecedented highness. For these businesses, organizations and groups, server downtime will cause serious damage, and this may even cause paralysis of the entire business operation. Because of the above reasons, the highly available technology came into being. The main purpose of highly available systems is to ensure that the commercial system keeps continuous service and high reliability, improve user experience and provides the high competitiveness for the enterprise in the technology layer.The main content of this paper is to provide stably and highly reliable system solutions for a distributed service under the cluster system. The paper analyzes the key points to achieve high availability systems, and make an analysis of need for high-availability system. The paper refers to a large number of high-availability data and summarizes previous work about achieving high availability systems, analyze its strengths and weaknesses, at the same time, analysis and study the key technology of the high availability in depth, make a summary of design for the highly available system based on the services and large data. Design each system module of highly availability in detail.Implement server scheduling module, data synchronous section and daemon modules, etc.by coding, and test the key components of highly available systems, heartbeat and dual hot simulation, in experimental environment.The highly available system uses advanced and mature technology, highlights heartbeat control and the principle of data storage solutions as well as method. Merge the two points on the basis of previous work.The main work of this paper as follows: 1. Review previous work, make a generalization of the previous design and realization of high-availability systems, analyze the key modules of the system, and point out the advantages and disadvantages of the system.2. Study and discusses the concept of high availability, as well as design points, based on services and data high availability system.Put forward some question should be noted when design the highly available system.3.Design the overall framework of the highly available system, separate system services and data storage, and give the specific design Photos of each module.4. Double parts of highly available systems:high availability of data and high availability of service. Give the design Photos about each part. Highly available module of data uses active and standby mode, monitor status, achieve double switch. A data server manages multiple servers which provide server actually. Custom protocols ensure communication between data server and the actual server in application layer. Set daemon on the actual server, it can immediately start the server which stops.5.The detailed design about highly available data server:The module carry out data asynchronous replication from master data server node to slave date server node by mysql’s binary log.Execute the interception of log using crontab in the main and slave data server on the definite time, send and import database operations to achieve the data remote disaster.6.Make a configuration on the key system’s modules.Preparing lab environment to test and verify the high availability of the system.The main innovation of this paper:By a combination of two high-availability architecture, separate data storage and server. Storage server with multi-node deployment use the way of a node and a backup to ensure high availability of the data server. The actual server uses the multi-machine mode. The custom protocol on application layer ensures communication between the two servers.Carry out the task scheduling module.The storage server accept client requests timely to check the actual server status, query the actual server status in real time, return server address in real time, ensure the high availability of the service. |
PDF Full Text Request