4. Relationship between hadoop + hbase + zookeeper

There are several main relationships:

1. the results produced after map and reduce operations seem to have been written to hbase. However, when the files in hlog and storefile of hbase are flush to disk, these two files are stored in the HDFS datanode, and HDFS is permanently stored.

2. What is the relationship between zookeeper and hadoop core and hbase? What services does zookeeper provide? They manage namenode in hadoop clusters, hbasemaster election in hbase, and status synchronization between servers. Specifically, a single zookeeper instance in hbase is responsible for the storage of hbase schema, real-time monitoring of hregionserver, and storage of addressing entries for all region, of course, the most common function is to ensure that there is only one master in the hbase cluster.