Improving Data Archival Performance of HDFS in Hadoop by E-PABR Algorithm

Hadoop Distributed file system (HDFS) is highly a fault-tolerant distributed file system associated with Hadoop framework. HDFS can handle a large amount of {data|of knowledge|of information} known as big data. HDFS deals with information archival as well. data archiving is a phenomenon which finds inactive data and moves it into a separate storage premise. Cloud-based storage facilitates it cost effectively while Hadoop clusters provide the computational power required. However, protecting archived data is the main concern of the information owner purpose of read. Erasure secret writing (EC) may be a methodology that has the mechanism to regain lost data as well. Of late a HDFS was developed to have special data archival features with (EC). The problem with it is that it needs similar the computational price for information of various sizes. Towards this finish, we proposed a methodology to beat this downside. A model application has built to exhibit evidence of the idea. The empirical results revealed that the methodology presented improves the computational efficiency in rendering data archival services.