HDFS stores all the data in terms of Blocks. With Block structure there are some benefits that HDFS gets. Some of these are as follows:
- Fault Tolerance: With Block structure, HDFS implements replication. By replicating same block in multiple locations, fault tolerance of the system increases. Even if some copy is not accessible, we can get the data from another copy.
- Large Files: We can store very large files that cannot be even stored one disk, in HDFS by using Block structure. We just divide the data of file in multiple Blocks. Each Block can be stored on same or different machines.
- Storage management: With Block storage it is easier for Hadoop nodes to calculate the data storage as well as perform optimization in the algorithm to minimize data transfer across the network.