The Enforcement of Security Policies for Computation
Abstract
Security policies define how information within a computer system is to be used. Protection mechanisms are built into these systems to enforce security policies. However, in most systems it is quite unclear what policies a particular mechanism can or does enforce. This paper precisely defines security policies and protection mechanisms in order to bridge the gap between them with the concept of soundness: whether a protection mechanism enforces a specific policy for a given program. Different sound protection mechanisms for the same policy and program can then be compared (on the basis of completeness) to determine if one outperforms the others. That a union of mechanisms for the same policy and program can be taken to produce a more complete mechanism is demonstrated. Although a maximal mechanism exists it cannot necessarily be effectively found. In addition to developing a theoretical framework in which to discuss security, a surveillance protection mechanism is introduced which indicates that it is sound and that it is more complete than the commonly used high water mark mechanism.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 1975
- Accession Number
- ADA013114
Entities
People
- Anita K. Jones
- Richard J. Lipton
Organizations
- Carnegie Mellon University