Multi-split HDFS Technique for Improving Data Confidentiality in Big Data Replication

Abstract

In this paper, the Secure Distributed Redundant Independent Files (SDRIF) approach addresses some issues found with the Redundant Independent Files (RIF) approach and it mainly introduced data confidentiality that the RIF approach lacks to offer. It works similar to RIF, but the generated parity is not stored in one separate file. The generated parity blocks are distributed among all four data parts. The CPSDRIF, which is the model produced when combining SDRIF with cloud providers (CP), introduces data security through the multi-split HDFS technique by distributing the parity blocks and reducing the size of the SDRIF block to the HDFS block. According to the experimental results to the CPSDRIF system using the TeraGen benchmark, it is found that the data confidentiality using CPSDRIF have been improved as compared to CPRIF. Also, the storage space is reduced by 33.3% with CPSDRIF system compared to other models and improved the data writing by 32% and reading by about 31%.