摘要
Hadoop是一个由Java语言实现的软件框架,在一个计算机集群中对海量数据进行分布式计算,这个集群可以支持上千个节点和PB级别的数据.HDFS是专门为Hadoop设计的分布式文件系统,作为Hadoop最基本的构成元素,保证内容的完整性与可用性.但是HDFS的操作界面并不友好,必须通过命令行或者IDE插件来实现对文件的操作.在研究HDFS的架构的基础上,设计实现了基于WEB的HDFS可视化文件访问系统.
Hadoop is a software framework realized with Java.It is designed for distributed computation in a computer cluster for massive data.This cluster can support thousands of nodes and PB level data.HDFS is designed specifically for the Hadoop as a distributed file system.As the basic constituent element of Hadoop,HDFS ensures the integrity and availability of the files in Hadoop.However,the interface of HDFS is not friendly.Users have to use command line or IDE plug-ins to achieve operation of the document.This paper studied the architecture of HDFS and then designed a HDFS visualization system based on WEB interface.
出处
《天津理工大学学报》
2012年第1期31-34,共4页
Journal of Tianjin University of Technology
基金
国家自然科学基金(61170301)
天津市科技支撑计划重点项目(08ZCKFGX00600)
天津市信息化项目(091052012)
