Damage Assessment and Recovery

Funy dependency in a database delineates a loose dependency relationship between two sets of attributes. It describes logical relationships among attributes in a database relation and those relationships can't be fully specified by functional dependencies, which focus on database schema and data organization. This characteristic of the databa se schema can be used to perform damage assessment and also to build fUz:.y recovery modeL In this paper, we formally define the concept offuz:.y dependency and introduce several inference rules. Then we focus on recovery from information attacks. An architecture for fuUJ' value generation during recovery, based on fuw dependency relationships, is also presented. FuUJ' dependency can accelerate the post attack recovery process because it can geflerate acceptable values for damaged data quicker compared to that in traditional recovery schemes. Although the generated fuzzy values may not offer the absolute accuracy, they are acceptable for many real-time applications, which require prompt response more than the data accuracy.

[1]  Anish Arora,et al.  Designing Masking Fault-Tolerance via Nonmasking Fault-Tolerance , 1998, IEEE Trans. Software Eng..

[2]  Sushil Jajodia,et al.  Surviving information warfare attacks on databases , 1997, Proceedings. 1997 IEEE Symposium on Security and Privacy (Cat. No.97CB36097).

[3]  Peng Liu,et al.  Efficient damage assessment and repair in resilient distributed database systems , 2001 .

[4]  Yair Amir,et al.  From total order to database replication , 2002, Proceedings 22nd International Conference on Distributed Computing Systems.

[5]  Ramez Elmasri,et al.  Fundamentals of Database Systems , 1989 .

[6]  Brajendra Panda,et al.  An architecture for making data available ceaselessly during recovery , 2003, IEEE Systems, Man and Cybernetics SocietyInformation Assurance Workshop, 2003..