Data management
We discussed HDFS blocks and replication in the previous sections. NameNode stores all metadata information and is a single point of failure, which means that no one can use HDFS if NameNode is down. This metadata information is important and can be used to restart NameNode on other machines. Thus, it is important to take multiple backup copies of a metadata file so that, even if metadata is lost from the primary NameNode, the backup copy can be used to restart the NameNode on the same machine or another machine. In this section, we will discuss NameNode metadata files such as fsimage and edit log. We will discuss data integrity further by using checksum and taking snapshots of the directory to avoid data loss and modification. ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access