Constructing a Secure MapReduce Framework in the Cloud Environment

MapReduce is a programing paradigm for large-scale data processing. Many cloud vendors are offering MapReduce service on their clouds so that the cloud user could deploy their MapReduce clusters on demand. However, in the current business model, the entire MapReduce cluster is deployed on a cloud and the MapReduce computation is opaque to the customer. Under this model, the attackers inside or outside of cloud may compromise the computation integrity and confidentiality without the knowledge of cloud users. From the integrity perspective, merely one compromised virtual machine in the MapReduce cluster can render the overall computation result inaccurate. From the confidentiality perspective, the malicious attacker or the abusive cloud insider could reverse-engineer the outsourced MapReduce jobs, in order to detect the logic or the algorithm of the outsourced job and therefore compromise the intellectual property of the cloud customer.

The aim of this project is to address two problems: 1) How to offer MapReduce computation with high result integrity on a public cloud environment that contains malicious nodes? 2) How to protect the confidentiality of control flow in the MapReduce job outsourced by the cloud user?


Papers:

  1. Yongzhi Wang, Jinpeng Wei. "Toward Protecting Control Flow Confidentiality in Cloud-based Computation". Computers & Security, Elsevier Ltd., accepted. Poster (468 KB).
  2. Yongzhi Wang, Jinpeng Wei, and Yucong Duan. "Securing MapReduce Result Integrity via Verification-based Integrity Assurance Framework". International Journal of Grid and Distributed Computing (IJGDC), Vol. 7, No. 6 (2014), pp. 53-70. Full paper (733 KB).
  3. Yongzhi Wang, Jinpeng Wei, Mudhakar Srivatsa. "Cross Cloud MapReduce: A Result Integrity Check Framework on Hybrid Clouds". International Journal of Cloud Computing (IJCC), ISSN 2326-7550, Vol. 1, No. 1, pages 26-39, July-September, 2013. Full paper (513 KB).
  4. Yongzhi Wang, Jinpeng Wei, Mudhakar Srivatsa, Yucong Duan, and Wencai Du. "IntegrityMR: Integrity Assurance Framework for Big Data Analytics and Management Applications". Proceedings of the IEEE International Conference on Big Data (IEEE BigData 2013), October 6-9, 2013, pages 33-40. Full paper (1131 KB), Slides (9364 KB).
  5. Yongzhi Wang, Jinpeng Wei, and Mudhakar Srivatsa. "Result Integrity Check for MapReduce Computation on Hybrid Clouds". Proceedings of the 6th IEEE International Conference on Cloud Computing (IEEE CLOUD 2013), IEEE Computer Society, Washington, DC, June 27-July 2, 2013, pages 847-854. Acceptance rate: 25%. Full paper (548 KB), Slides (657 KB), Poster (1.01 MB).
  6. Yongzhi Wang, Jinpeng Wei, and Mudhakar Srivatsa. "Cross Cloud MapReduce: an Uncheatable MapReduce". The 33rd IEEE Symposium on Security and Privacy, San Francisco, CA, May 20-23, 2012. Poster paper. Paper (582 KB), Poster (521 KB).
  7. Yongzhi Wang, Jinpeng Wei. "VIAF: Verification-based Integrity Assurance Framework for MapReduce". The Fourth IEEE International Conference on Cloud Computing (CLOUD 2011), July 4-9, 2011, Washington DC. Full paper (376 KB), Slides (683 KB).