HDFS (Hadoop Distributed File System) is one of the component of the Hadoop framework.
HDFS is a distributed file system, which means that it is using a cluster of servers to create a storage space.
HDFS can store any number of files. The files stored in the cluster are divided into data blocks and distributed among the different nodes of the HDFS cluster.
To guarantee data resilience, each data block is duplicated on several nodes (three by default).
Therefore, the file system is not affected if a node fails.

To handle files on a HDFS server, you can:

1. Connect to the HDFS server (hdfsOpenConnection).
2. Perform one of the following operations:
   • copy files from the local computer onto the server (hdfsCopyTo).
   • copy files from the server onto the local computer (hdfsCopyFrom).
   • create a directory on the server (hdfsCreateDirectory).
   • rename a file on the server (hdfsRename).
   • delete a file from the server (hdfsDelete).
   • list the content of a directory on the server (hdfsList).
3. Close the connection to the server (hdfsCloseConnection).

Caution: The HDFS functions are available in edit and at run time from Windows Vista or Windows Server 2008.