Why Hbase doesn't use all nodes in the cluster? - hadoop

Question

I'm setting up a Hbase cluster and I ran into a problem. When I'm writing my data to the cluster some nodes remain empty.
Hbase Status Screen :
Dfshealth screen :
Hbase 1.4.10, Hadoop 3.1.2
node-master hbase-site.xml
<configuration>
<property>
<name>hbase.rootdir</name>
<value>hdfs://node-master:9000/hbase</value>
</property>
<property>
<name>hbase.zookeeper.property.dataDir</name>
<value>hdfs://node-master:9000/zookeeper</value>
</property>
<property>
<name>hbase.cluster.distributed</name>
<value>true</value>
</property>
<property>
<name>hbase.zookeper.quorum</name>
<value>node-master</value>
</property>
<property>
<name>hbase.zookeeper.property.clientPort</name>
<value>2181</value>
</property>
<property>
<name>hbase.client.write.buffer</name>
<value>8388608</value>
</property>
<property>
<name>hbase.client.scanner.caching</name>
<value>10000</value>
</property>
</configuration>
node-master regionservers (hadoop workers same)
node1
node2
node3
node4
node5
node6

Answer 1

Hbase writes data in regions where each region is a collection of rows based on sorted keys. When you reach a region limit HBase split the region into 2 regions and same when 2nd region reach limit it again split. each region is assigned to a region server(Datanode).
So your table does not have as many regions that it will utilize all the nodes. so to balance the data across nodes you can either pre-split the table while creating it.
Hbase Table Pre-split
Please also read about HBase hotspot problem then you get more understanding.