We focus on disk reads rather than disk writes, because the majority of disk accesses are for read operations, and the majority of disk bandwidth is used for read operations.
我们关注的是磁盘读取,而不是磁盘写,因为大部分磁盘访问都是为了进行读取操作,且大部分磁盘带宽也用于读取操作。
So, the user can read several article synopsis items on one page, and the server bandwidth is spared from building huge pages even if the user doesn't read it all.
因此,用户可以在一个页面中同时阅读几篇文章的概要内容,并且构建大量的页面会使服务的带宽不够,虽然用户并不会同时访问所有的页面。
To minimize global bandwidth consumption and read latency, HDFS tries to satisfy a read request from a replica that is closest to the reader.
副本的选择,为了降低整体的带宽消耗和读延时,HDFS会尽量让reader读最近的副本。
应用推荐