HDFS goals
Fault detection and recovery : Because HDFS contains a large number of commodity hardware, the probability of failure of components is very high. Therefore, HDFS has to have mechanisms to identify and recover quickly and automatically.
The huge data set : HDFS has hundreds of nodes in each cluster to manage applications with large data sets.
Hardware in Data - When calculations occur near data, a task can be done effectively. Especially when large data sets are involved, it reduces network traffic and increases performance.