The Research Of Data Memory Caching Technology On Recommended Platform For Scientific Literature

With the rapid development of the Internet, Massive data and high concurrency have become the major performance bottleneck of the PC and WAP applications. In order to solve this problem, the caching technology is applied. A good cache design can reduce the pressure of the server and reduce the response time of the application. The caching, especially the distributed caching, has very remarkable effect to solve this performance bottleneck. This paper is to research the data memory caching technology.By this way, the paper can meet the access performance requirements of the recommended platform for scientific literature, which has massive data and high concurrency. By using data memory caching, the platform reduces the response time and improves the user's experience.This paper focus on data caching technology research, and apply this technology to recommended platform for scientific literature. The main works are as follows:First, the data caching technology, especially the distributed caching technology is studied. Redis and Memcached, which are commonly used in the distributed caching, are compared from their own characteristics and performance under the platform. Finally Redis is chosen for distributed caching. And the existing distributed caching structers are studied. and finally the paper chooses the proxy middleware Twemproxy to build the distributed caching for the platform.Second, the distributed caching is optimized. This paper adjusts the source code parameters to optimize the distributed caching based on the source code research and the specific business.Third, design and implement the recommended platform for scientific literature based on data caching. And the function modules and data storage modules are improved and realized, and the platform availability is improved.The recommended platform for scientific literature is stable which is based on the research results of this paper. The response time of the platform is greately reduced by building distributed caching system, and the cache memory is greately reduced.