Registration

Local/Hotel Information

Detailed Session Information

Tutorials

Keynote Speaker

DSN Invitation

Workshops

PoDSy 2003: Principles of Dependable Systems

Dependable systems are supposed to satisfy an ensemble of distinct properties, namely safety, security and availability, to name a few. These properties are in parts complementary and also diverse enough to have spawned complete topic areas of their own. Consequently, work on achieving and validating the different properties has partly been performed in different communities and with varied nuances. Maybe most prominently this is true for the two areas of fault-tolerant systems on the one hand and secure systems (especially cryptography) on the other. For example, researchers in fault-tolerance often make statements about systems by treating cryptographic primitives as black boxes. This is done to simplify analysis and (sometimes) avoid number and probability theory. However, by abstracting away the basic properties of the cryptographic primitives, this severely constrains the ability to conduct rigorous security proofs. Various examples of the past show that by over-abstraction, important attributes got neglected, contributing to attack vulnerabilities in the resultant protocols. But despite these examples, many researchers have confirmed that there are strong similarities between the ways of modeling and handling uncertainty in both areas.

This workshop brings together researchers and practitioners from both the fault-tolerance and security communities to discuss foundational topics (and related applied experiences) on the similarities and differences between both areas. The workshop is structured around a set of invited talks by well-known specialists which have experiences in both areas. It is complemented by a panel which explores the question, what fault-tolerance people can learn from security people and vice versa.

Invited speakers: Paulo Verissimo, University of Lisboa; Ran Canetti, IBM Research; Catherine Meadows, Naval Research Laboratory; John Knight, University of Virginia

Research Papers

Panel: "What can fault-tolerance people learn from security people and vice versa?", Neeraj Suri, TU Darmstadt, Leslie Lamport, Microsoft Research, Jonathan Millen, SRI International, Yves Deswarte, LAAS, Roy Maxion, Carnegie Mellon University

Co-Chairs: Felix Gartner (EPF, Switzerland), Klaus Kurawe (IBM, Switzerland), Levente Buttyan (Budapest University of Technology and Economics, Hungary)

Model Checking for Dependable Software-Intensive Systems

Model checking is indispensable in the development of modern digital circuitry and is emerging as a valuable instrument for software verification. Model checking has uncovered errors in a variety of software-intensive systems, including spacecraft redundancy management, aircraft collision avoidance, and weapons control systems. The approach offers the potential to help ensure behavioral properties and eliminate catastrophic errors in software systems that require high levels of dependability.

While it offers significant promise for dependable system analysis, model checking faces a variety of technical and practical challenges that must be overcome. These range from theoretical questions relating to optimal analysis approaches, to implementation constraints like state explosion, to adoption issues faced by a development organization.

This full-day workshop will be a forum for sharing research results, practical experiences, and advances in the application of model checking as a software engineering tool. All of the sessions of the workshop will be interactive. In these sessions, attendees and presenters will discuss, identify, and detail the state of the technology and the technical advantages, limitations, and critical research areas impacting the effective use of model checking in dependable software-intensive systems design and analysis.

Co-Chairs: Edmund Clarke (CMU, USA), Masahiro Fujita (University of Tokyo, Japan), David Gluch (Embry-Riddle University, USA)

First Workshop on the Design of Self-Managing Systems

See workshop webpages at http://www.cs.stanford.edu/~candea/dsms.html

As systems become increasingly connected to an increasingly diverse set of other systems and environments, architects will lose their ability to intricately plan interactions among system components, because an increasing fraction of those interactions will be with foreign and possibly unanticipated systems or components. Humans will be increasingly less competent to install, configure, optimize, maintain, and merge massive, complex, and heterogeneous computing systems. They will not be able to make sufficiently quick, decisive responses to a rapid stream of changing and conflicting demands.

To meet these new and vast challenges, systems need to be designed so that they are self-managing. Self-managing systems automate all phases in the life-cycle of complex computing systems, from installation to run-time maintenance and optimization. This workshop will provide a forum for a small group of participants to consider new paradigms, system designs, and algorithms for self-managing computing systems.

We solicit papers addressing topics that include, but are not limited to: self-managing storage, peer-to-peer architectures, and grid architectures

• economic models for self-management
• biological models for self-management
• self-managing transaction systems
• adaptation to human errors
• limits and dangers of relying on self-managing systems
• decision algorithms
• data-mining, statistics, and other analytic techniques

Workshop attendance is open to all attendees registered for DSN-2003. We welcome participation by professionals with diverse backgrounds, who can contribute to advancing the technology and understanding of the workshop subject.

Co-Chairs: Lisa Spainhower (IBM, USA), Aad van Moorsel (HP, USA)

SSS'03: Sixth Symposium on Self-Stabilizing Systems

Self-stabilization contrasts with other approaches to dependable systems -- instead of entirely masking failures by replication or safe storage, or using backward recovery to restore system state, self-stabilization is pure forward recovery, tolerating any degree of transient failure. Though self-stabilization has been a standard algorithmic topic of distributed systems, there is now renewed interest due to emerging concerns of scalability and more extreme distribution of system control. There is also recognition that system components become more valuable if they are individually self-stabilizing.

Meetings of researchers interested in self-stabilization have been stimulating and useful. At recent workshops, new adaptations of stabilization have been proposed, new technical advances in algorithms and analysis have been discussed, and perhaps most important, new problem areas and applications suited to the stabilization approach became known to a wider community. This workshop will continue these themes with high-quality research presentations and open discussion of the latest results and most exciting new directions.

Tentative Schedule:

There will be 15 to 18 technical presentations, and open discussion session, and perhaps some short presentations of recent work of interest to the community of researchers on self-stabilization.

Invited Talk: Anish Arora, Ohio State University

Chair: Shing-Tsaan Huang (National Central University, Taiwan)