HDFS (HaDoop File System)

HDFS is one of utils that included at Hadoop framework.

• HDFS knows to handle large file (Ex. 1ptb) and split it to Data blocks and distrebute it over Cluster.
• HDFS has Fault tolerance - HDFS replicate same Data block on X Data nodes. X is Replication factor
• HDFS Master / Slaves architecture

• Rack awarness - What happends if entire rack is lost ?
• Reliable storage
• High throughput

• dfsadmin
• fs shell
• Web-UIs

• Name node - FS Ops, Block Mapping
• Secondary name node - Checkpoint Ops
• Data node - Block Ops, Replications

See: MapReduce

• Master: Job tracker - controller of Task trackers
• Task tracker