Hadoop framework provides a facility to store large and large amounts of data with almost no breakdown while querying. It breaks the file into pieces, copies it multiple times (3 default) and stores it on different machines. Accessibility is ensured even if any machine breaks down or is thrown out from network.
One can use Map Reduce programs to access and manipulate the data. The developer need not worry where the data is stored, he/she can reference the data from a single view provided from the Master Node which stores all metadata of all the files stored across the cluster.
0 comments:
Post a Comment