Privacy-Preserving Techniques for Big Data Analysis in Cloud
Big Data is termed as the enormous collection of large datasets that cannot be processed by using the traditional computational technique. Big data consists of different formats like structured data and unstructured data; it is not efficient in terms of response time to perform analysis of big data using existing traditional RDBMS techniques. The Big data possess many challenges because of several properties like variety, value, variability, volume, visualization, velocity and veracity. Big data is not allowed to store on one machine due to its collection of enormous datasets so it requires a new distributed platform to analyze and store such huge amount of datasets for future prediction and decision making. Hadoop provides the framework for distributed data processing to analyze and store this large amount of data using commodity hardware. Cloud computing can process the distributed queries across multiple datasets with the use of Hadoop. Big data is stored in the cloud computing environment for processing and analyzing the enormous datasets. Because of this paradigm, many new challenges for securing userÔ??s sensitive data, privacy-preserving mechanisms are used in untrusted cloud servers. Privacy-preserving mechanism cryptographically enforced the security to the private data.In this paper, to provide data confidentiality and integrity, three techniques Homomorphic encryption, order-preserving encryption schemes and Attributebased encryption are described. These techniques can preferably use in the cloud to provide data privacy and also preferred for big data to maintain the efficiency and scalability of large datasets for decision making.
MapReduce, Big data, HDFS (Hadoop distributed file system), Cloud computing, Hadoop, Homomorphic encryption (HE) scheme, order-preserving encryption (OPEs), Attribute-based Encryption (ABE).