Hadoop over IBM Storage Scale

By default, Hadoop uses HDFS schema (hdfs://<namenode>:<portnumber>) for all the components to read data from HDFS or write data into HDFS.

Hadoop also supports local file schema (file:///) and other HCFS schema for Hadoop components to read data from or write data into other distributed file systems.

IBM Storage Scale HDFS Transparency follows the Hadoop HCFS specification and provide the HDFS RPC level implementation for Hadoop components to read data from or write data into IBM Storage Scale. It also supports Kerberos, federation, and distcp.