The HBase-HDFS service is built on top of Hadoop Distributed File System (HDFS). HBase-HDFS is used to store WAL files of HBase and ensure the persistence and high reliability of HBase logs.
HBase-HDFS
HDFS is a core component of the Hadoop ecosystem that stores distributed files in a reliable manner. HBase-HDFS uses HDFS to store underlying data. HBase-HDFS inherits all native features and benefits of HDFS, without the basic architecture of HDFS modified. For more information about HDFS, see HDFS overview.
HBase-HDFS is automatically deployed in the following cases:
If you select the OSS-HDFS and HBase services and select the check box for HBase Log Storage when you create an E-MapReduce (EMR) cluster, the HBase-HDFS service is deployed after the cluster is created.
If you select the OSS-HDFS service when you create an EMR cluster and add the HBase service after the cluster is created, the HBase-HDFS service is deployed.
References
For information about how to view the deployment information of service components, see View the deployment information of service components.
For information about how to unpublish an HBase-HDFS DataNode, see Scale in a node group.