Research On And Application Of Data Caching Method Based On MongoDB And Redis

The combination between the development of Internet technology and software technology brings the multi-tenant applications, which is very flexible and holds huge numbers of users. However, with the increasing numbers of tenants settled on the multi-tenant applications, many problems occurred in the data layer of applications. In order to meet the new requirements of multi-tenant applications and ensure high-speed data access responses, the data structure of data-layer needs to be able to guarantee high scalability and configurability. When many tenants are trying to access the data at the same time, high-speed data access can ensure the applications' quality of service.From the perspective of the development of multi-tenant applications, this thesis study the cache prediction and cache replacement issues occurred in using MongoDB, selecting the high performance in-memory database Redis as the cache container. After studying the multi-tenant applications' data cache requirement features, the thesis focuses on the following aspects:(1) Establish a multi-tenant application cache management data model, and implement the configuration, generation and parsing of multi-factor expression in the model. Analyze the features of multi-application, multi-tenant data selection, and describe the selection of cache choosing conditions by factors. Apply the factor expression to the cache selection.(2) Analyze the feasibility of whether a tenant can select a Redis cache replacement strategy. By analyzing the selectable hierarchical structure of Redis cache replacement strategy, the necessary conditions for tenants to choose cache policy are obtained. Based on the study of the realization of Redis replacement strategies, design and implement three kinds of commonly used cache replacement algorithms in order to meet different scenarios and different replacement requirements. The data structures used in implementing Redis cache replacement algorithms can also provide the reference for users who are unfamiliar with Redis database.(3) Define the application interface of multi-factor-based cache selection for multi-application and multi-tenant systems, as well as the interface of Redis replacement policy configuration. By using it in practical applications, its effectiveness has been verified.