Security Theories and Practices for Big Data

Abstract

Big data applications usually require flexible and scalable infrastructure for efficient processing. Cloud computing satisfies these requirements very well and has been widely adopted to provide big data services. However, outsourcing and resource sharing features of cloud computing lead to security concerns when applied to big data applications, e.g., confidentiality of data/program, and integrity of the processing procedure. On the other hand, when cloud owns the data and provides analytic service, data privacy also becomes a challenge. Security concerns and pressing demand for adopting big data technology together motivate the development of a special class of security technologies for safe big data processing in cloud environment. These approaches are roughly divided into two categories: designing new algorithms with unique security features and developing security enhanced systems to protect big data applications. In this chapter, we review the approaches for secure big data processing from both categories, evaluate and compare these technologies from different perspectives, and present a general outlook on the current state of research and development in the field of security theories for big data.