Research And Application Of Distributed Data Caching Technology

With the popularity of the Internet, more and more users get information and services via the Internet. When the modern Internet Applications of Enterprise face a large number of requests, usually using data cache system to reduce waiting time of the users'requests and the load of the database server. Data cache system can effectively reduce the accessing load of the database server, reduce waiting time of the users'requests, and improve the overall system's performance of application. In recent years, there have been many open-source cache system, but they all have a shortage in data redundancy backup and failure recovery, while commercial cache system is not only costly, but also opaque, and it can't be used most efficiently. Therefore, this paper designs and implements a Distributed Data Cache System which has the function of data redundancy backup and failure recovery, according to the actual needs of some website of inquiring air tickets.First, the paper studies distributed caching technology in depth, specifically including cache topology structure, data distribution mode, cache synchronization strategy, cache replacement algorithm and distributed caching protocol. Then, the paper analyses important parts of distributed data cache system in depth, designs a data distribution mode based on consistent hashing algorithm with virtual nodes, and in the data distribution mode develops a set of communication protocols between cache nodes based on JGroups technology. Next, the paper analyses traditional cache replacement algorithm and the factors that affect cache performance, and designs a new cache replacement algorithm used for caching airline ticket data. The algorithm has a higher request hit rate and byte hit rate through simulating logs. In the end, based on researching on distributed caching technology, the paper designs and implements the cache system which has the function of data redundancy backup and failure recovery. The system uses Peer-To-Peer topology structure, and includes cache management module, data distribution module, cache synchronization module, replacement algorithm module, cache communication module and reliability service module. By test, the cache system provides reliable mechanisms of data redundancy backup and failure recovery, and it runs stably.