基于面向对象的大数据存储模型研究

摘要

Along with the continuous development of the internet of things and cloud computing application,the storage and management of massive data become more and more important. Traditional data storage methods can ’t meet the needs of the big data storage,and the industry provides various kinds of big data storage models. The paper compares and analyzes various storage models of big data,and provides an object oriented storage idea based on data character virtual types and object. Then a big data storage model is designed,and testing and analysis is made to the model classification key model and non-structured data storage performance. The testing result disclosed that the model classification method has less mistakes,and the non-structured data storage performance is higher. The model has good validation and availability and provides direction for big data high efficient storage and intelligent management research.%随着大数据存储需求的不断扩大，网络存储技术面临如何存储并管理海量数据的问题。通过对现有各种大数据存储模型进行了对比分析，针对现有存储模型存储的局限性和大数据存储的特点，提出了一种基于数据特征的面向对象存储思想。采用虚拟类技术设计并实现了一种基于面向对象的大数据存储模型。在仿真环境中，对该模型的分类关键模块与非结构化数据存储性能进行了测试与分析，实验结果表明该模型分类模块误差较小，读写效率较高，而且随着数据的增大，非结构化存储性能保持稳定。