Insights into Composability from Lablet Research

Jonathan Aldrich, William L. Scherlis, Anupam Datta, David Garlan, Bradley Schmerl, Joshua Sunshine, Christian Kaestner, André Platzer , Limin Jia, Robert Harper, Travis Breaux, Witawas Srisa-an, and Arbob Ahman, Carnegie Mellon University.

This presentation by Jonathan Aldrich from the CMU Lablet described a framework for understanding the hard problem of composability in the setting of security, along with highlights of lablet research results illustrating recent progress in this area and remaining research challenges. The format was an open discussion, and it proved lively.

Prof. Aldrich identified the primary challenge as the need to develop methods to construct secure systems with known security properties from components each of which has known quality and security properties, and avoid full reanalysis of the constituent components. Composition is needed to manage Increasing scale, complexity, dynamism, socio-technical ecosystems, and rich supply chains, and to direct evaluation of artifacts as they are produced and evolved.

The CMU SoS Lablet approach has been to focus on the hardest technical problems, emphasizing composability of modeling and reasoning as a key to scale and incrementality and human behavior and usability for developers, evaluators, operators, and end users. From this work, they seek to advance scientific coherence of cybersecurity technical results, advance most-effective scientific processes, acknowledge the multidisciplinary nature of cybersecurity, enhance the coherence of the body of technical results, enhance productivity, validity, and translation into practice and engage and broaden the cybersecurity technical community. To expand the community, they facilitate community and educational engagement with subcontractor partners, workshops, and conference events.

Work to date includes an initial workshop held in September, 2013. At this workshop, they developed a series of definitions, issues and approaches including crosscutting principles using assume-guarantee reasoning, game theory, and families of systems.

One key element was utilizing the work on sequential compositionality by Ahmad and Harper that produced the logical statement: if two components preserve confidentiality and we compose them in sequence, then the result preserves confidentiality. This premise underlies compositional security and is stated as: If two components preserve confidentiality and we compose them in sequence, then the result preserves confidentiality.

The PowerPoint presentation is available at: http://cps-vo.org/node/15746

Prof. Aldrich’s personal web page is available at: http://www.cs.cmu.edu/~aldrich/

(ID#:14-2626)

Note:

Articles listed on these pages have been found on publicly available internet pages and are cited with links to those pages. Some of the information included herein has been reprinted with permission from the authors or data repositories. Direct any requests via Email to SoS.Project (at) SecureDataBank.net for removal of the links or modifications to specific citations. Please include the ID# of the specific citation in your correspondence.