Computation Integrity in Distributed Volunteer Computing Platforms Papers -  People -  Talks

Research Summary

Poster for CyberTrust PI Meeting January 2007

This research investigates the protection of computation integrity and enhancement of data privacy for applications executed on distributed computing platforms that utilize the spare processor cycles of computers connected to the Internet. The goal of the project is to provide computation supervisors with quantifiable protection and privacy mechanisms tailored to application and platform specific constraints. The project considers existing platforms, which prohibit participating computers ("participants") from communicating with other participants, as well as potential future hierarchical platforms that permit communication among participants and allow dynamic variation of platform topology. The approach to the problem of computation integrity consists of probabilistic verification and novel application of artificial intelligence mechanisms for detecting anomalous behavior. The approach to data privacy is based on obscuring data values while maintaining sufficient output information for identifying specific important data from among a vast data set. The experimental research links to education at the University of Richmond through development of special topics courses and enhancement of existing courses, and by engaging undergraduates in relevant research via funding of research fellowships, equipment, and travel to project related conferences and workshops. The results of the project will provide access to more secure computing platforms, fostering inter-disciplinary research among scientists and practitioners. Specific fields benefiting from results of this research include biology, chemistry, math, finance, medicine, and evolutionary theory. Project implementations will modify existing open source distributed computing code and will be disseminated via this project web site.

People

Current Students

Stephan Chipilov
Garret Graham
Ivan Jibaja
Matt King
Mike Pohl
Andy White
Brittany Williams

Former Students

Rom Chan
Trin Chavalittumrong
Robin Haskins Givens
Ed Kenney
Greg Steffensen
Dan Upton

Papers

• D. Szajda, M. Pohl, , J. Owen, and B. Lawson, Toward a Practical Data Privacy Scheme for a Distributed Implementation of the Smith-Waterman Genome Sequence Comparison Algorithm, Proceedings of the 2006 ISOC Network and Distributed System Security Symposium (NDSS 2006), pages 253-265, San Diego, CA, February 2006.
  
• D. Szajda, B. Lawson, and J. Owen, Toward An Optimal Redundancy Strategy for Distributed Computations, Proceedings of the 2005 IEEE International Conference on Cluster Computing (Cluster 2005), Boston, MA, September 2005.
  
• D. Szajda, J. Owen, B. Lawson, A. Charlesworth, and E. Kenney, An Alternate Multiplicity-2 Task Assignment Scheme for Distributed Computations. In Scheduling and Resource Management for Parallel and Distributed Systems (SRMPDS 05) in conjunction with The 2005 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA 05), pages 59-65, Las Vegas, NV, June 2005.
  
• D. Szajda, B. Lawson, and J. Owen, Hardening functions for large-scale distributed computations, Proceedings of the 2003 IEEE Symposium on Security and Privacy, pages 216-224, Berkeley, CA, May 2003.
  

Talks

• How to Tell a Secret Without Revealing It (Enhanced Data Privacy in a Distributed Implementation of the Smith-Waterman Genome Sequence Comparison Algorithm) (PPT)(Barry Lawson), presentation given at various undergraduate institutions in Spring 2006.


• Toward a Practical Data Privacy Scheme for a Distributed Implementation of the Smith-Waterman Genome Sequence Comparison Algorithm (PPT) (Doug Szajda). Network and Distributed System Security Symposium, San Diego, CA. February 2006.
  
• Hardening Functions for Large-Scale Distributed Computations (PPT) (Doug Szajda). IEEE Security and Privacy Symposium, Berkeley, CA. May 2003.

Reading Group

University of Richmond's Security Reading Group meets most weeks Wednesday afternoon at 3:30. See http://oncampus.richmond.edu/~dszajda/classes/systems_seminar/main.html for an updated schedule and information about joining.

University of Richmond Department of Computer Science

Supported by the Cyber Trust program

Doug Szajda dszajda@richmond.edu