基于内存的分布式缓存系统设计

发布时间：2018-04-27 20:40

本文选题：分布式 + 内存　；参考：《华中师范大学》2012年硕士论文

【摘要】：随着互联网WEB2.0时代的到来,信息传播的广度和深度都达到了一个新的高度。传统的网站都是用数据库来进行数据的存储、检索和更新,但随着访问量的增加,将导致数据库服务器系统资源紧缺,查询操作响应变慢,网站用户体验变差。因此,开发缓存系统具有广泛的应用前景。本文根据应用中一部分检索数据库的操作具有逻辑相同,请求频繁的特点,结合内存读写速率快的特点,提出并设计了一款分布式内存缓存系统。该系统采用物理内存作为数据存储介质,使用客户端／服务器的软件结构,通过在客户端应用分布式哈希算法实现了分布式存储的功能,并开发了PHP及Python的编程接口,给应用开发提供了一套可供简单调用的高性能存储服务。本文主要介绍了分布式内存缓存系统的体系结构以及各个层次的具体实现。本设计主要分为以下四个方面的内容：在数据库检索方案性能有限,迫切需要提高网站浏览体验的前提下,调研并提出了分布式内存缓存系统的完整设计方案；介绍了作为系统存储单元的内存缓存系统的设计及主要功能模块；通过采用哈希算法的客户端来将请求分散到不同的内存缓存节点上,达到分布式的效果；最后根据前台开发的特点,实现了PHP及Python的调用接口。通过将缓存系统读性能和数据库检索的压力测试结果进行对比,表明该系统具有较好的读性能和实用性。
[Abstract]:With the advent of Internet WEB2.0 era, the breadth and depth of information dissemination has reached a new height. Traditional websites use databases to store, retrieve and update data, but with the increase of visits, the database server system resources will be scarce, the response of query operations will become slow, and the site user experience will become worse. Therefore, the development of cache system has a wide range of application prospects. In this paper, a distributed memory cache system is proposed and designed according to the characteristics of the same logic and frequent requests in part of the retrieval database in the application, combined with the characteristics of fast memory reading and writing rate. The system adopts physical memory as data storage medium, uses the software structure of client / server, realizes the function of distributed storage by applying distributed hash algorithm in client, and develops the programming interfaces of PHP and Python. It provides a set of high performance storage service for application development. This paper mainly introduces the architecture of distributed memory cache system and the implementation of each level. This design is mainly divided into the following four aspects: under the premise of limited performance of database retrieval scheme and urgent need to improve website browsing experience, a complete design scheme of distributed memory cache system is proposed; This paper introduces the design and main function modules of the memory cache system as the storage unit of the system, distributes the request to different memory cache nodes by using the client of the hash algorithm, and achieves the distributed effect. Finally, according to the characteristics of foreground development, the calling interface of PHP and Python is realized. By comparing the read performance of the cache system with the stress test results of database retrieval, it is shown that the system has good reading performance and practicability.
【学位授予单位】：华中师范大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP333;TP311.13

【参考文献】