Hadoop分布式文件系统(HDFS)被设计成适合运行在通用硬件(commodity hardware)上的分布式文件系统。它和现有的分布式文件系统有很多共同点。但同时,它和其他的分布式文件系统的区别也是很明显的。HDFS是一个高度容错性的系统,适合部署在廉价的机器上。HDFS能提供高吞吐量的数据访问,非常适合大规模数据集上的应用。HDFS放宽了一部分POSIX约束,来实现流式读取文件系统数据的目的。HDFS在最开始是作为Apache Nutch搜索引擎项目的基础架构而开发的。HDFS是Apache Hadoop Core项目的一部分。
HDFS有着高容错性(fault-tolerant)的特点,并且设计用来部署在低廉的(low-cost)硬件上。而且它提供高吞吐量(high throughput)来访问应用程序的数据,适合那些有着超大数据集(large data set)的应用程序。HDFS放宽了(relax)POSIX的要求(requirements)这样可以实现流的形式访问(streaming access)文件系统中的数据。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