Secure Storage and Data Sharing Scheme Using Private Blockchain-Based HDFS Data Storage for Cloud Computing

Abstract

– The storage of a vast quantity of data in the cloud, which is then delivered via the internet, enables Cloud Computing to make doing business easier by providing smooth access to the data and eliminating device compatibility limits. Data that is in transit, on the other hand, may be intercepted by a man-in-the-middle attack, a known plain text assault, a selected cypher text attack, a related key attack, or a pollution attack. Uploading data to a single cloud might, as a result, increase the likelihood that the secret data would be damaged. A distributed file system extensively used in huge data analysis for frameworks such as Hadoop is known as the Hadoop Distributed File System, more commonly referred to as HDFS. Because with HDFS, it is possible to manage enormous volumes of data while using standard hardware that is not very costly. On the other hand, HDFS has several security flaws that might be used for malicious purposes. This highlights how critical it is to implement stringent security measures to make it easier for users to share files inside Hadoop and to have a reliable system in place to validate the shared files' validity claims. The major focus of this article is to discuss our efforts to improve the security of HDFS by using an approach made possible by blockchain technology (hereafter referred to as BlockHDFS). To be more precise, the proposed BlockHDFS uses the Hyperledger Fabric platform, which was developed for business applications, to extract the most value possible from the data inside files to provide reliable data protection and traceability in HDFS. In the results section, the performance of AES is superior to that of other encryption algorithms because it ranges from 1.2 milliseconds to 1.9 milliseconds. In contrast, DES ranges from 1.3 milliseconds to 3.1 milliseconds, three milliseconds to 3.6 millimetres, RC2 milliseconds to 3.9 milliseconds, and RSA milliseconds to 1.4 milliseconds, with data sizes ranging from 910 kilos.