SecureMR: A Service Integrity Assurance Framework for MapReduce

MapReduce has become increasingly popular as a powerful parallel data processing model. To deploy MapReduce as a data processing service over open systems such as service oriented architecture, cloud computing, and volunteer computing, the authors must provide necessary security mechanisms to protect the integrity of MapReduce data processing services. In this paper, they present SecureMR, a practical service integrity assurance framework for MapReduce. SecureMR consists of five security components, which provide a set of practical security mechanisms that not only ensure MapReduce service integrity as well as to prevent replay and Denial Of Service (DoS) attacks, but also preserve the simplicity, applicability and scalability of MapReduce.