ACM Logo
http://www.acm.org/conferences/sac/sac2006/

Dependable and Adaptive Distributed Systems

DADS Track of the
21st ACM Symposium on Applied Computing

April 27, 2006
Bourgogne University
Dijon, France

 

Call for Participation

 

Applied Computing has been a primary gathering forum for applied computer scientists, computer engineers, software engineers, and application developers from around the world. SAC 2006 is sponsored by the ACM Special Interest Group on Applied Computing, and is hosted by the University of Bourgogne, Dijon, France.

We invite you to actively participate in the "Dependable and Adaptive Distributed Systems" track and discuss the presented approaches with the participants.

 

Track Focus

Distributed systems and databases are at the core of the information society and increasingly pervade many aspects of our daily lives. While mobility and pervasiveness require support for systems that adapt themselves to changing environments, the middleware infrastructures become more and more heterogeneous and complex. In addition, we can see an increasing demand for dependability of such systems, taking into account the software as well as the surrounding environment. Generally, adaptiveness can either satisfy a change in user requirements or seek to fulfill the same requirements in a changing system context and environment. In particular, adaptation is also a means to achieve dependability in a computing infrastructure with dynamically varying structure and properties. Fault tolerance can consequently be seen as a special case, where the adaptation seeks to overcome an otherwise negative effect of a change in the computing infrastructure that can be classified as a fault. However, dependability can not only be achieved by fault tolerance, but also by other means like fault avoidance (e.g. through formal methods).

Therefore, future middleware needs to support adaptiveness and dependability while maintaining scalability and mastering complexity. Still, software legacy must be integrated in a way, such that open and standardized interfaces support not only functional integration, but also a seamless integration of non-functional aspects. While generic approaches can never provide optimal solutions, the "end-to-end" argument (i.e. let the application do the work) is not appropriate either, as about 80% of the infrastructure code is very similar throughout a wide range of applications. Of course dependability and adaptiveness can not simply be added to a system like a plug-in module. Rather, databases and middleware need architectural principles and sound methodologies based on formal methods, as well as component container services, component interaction and composition standards, and possibly consideration of software aspects to help application developers to weave their functional code into a configurable distributed middleware instead of re-inventing the wheel each time. Moreover, service-oriented architectures need coordination in order to achieve dependability and can further benefit from context-aware approaches.

In software engineering, application developers need tools, common practices, architectural principles, descriptive configurability, middleware services, document design support, and aspect-oriented programming as well as architectural principles and sound methodologies (e.g. MDA) based on formal methods tackling the inherent complexity of software and ensuring the development of trustworthy systems. This gives evidence of an apparent interest of industry as well. Therefore, the focus of this track is on middleware support for dependability and adaptiveness of distributed software systems and service environments of any kind and on methods of designing and engineering them. (We use the term adaptive middleware to denote middleware that supports the development of adaptive systems, but we do not focus on reflective middleware.)

The track provides a forum for scientists and engineers in academia and industry to present and discuss their latest research findings on selected topics in dependable and adaptive distributed systems and complex services. The topics of particular interest include, but are not limited to:

  • Architectural and infrastructural principles for adaptive and dependable distributed systems.
  • Dependability in complex Web services, GRID-computing, and P2P-systems. Co-ordination and context-awareness (context-modeling).
  • Middleware support for reunification and reconciliation. Consideration of alternative techniques for dynamic configuration and/or reconfiguration.
  • Data replication strategies, interfaces, and standards. Interaction of distributed databases with middleware systems.
  • Group communication and group membership services in failure scenarios with network partitions.
  • Support for dependability and adaptiveness in component-based systems (e.g. component frameworks, container services, deployment, composition and substitution of components, building trusted systems from untrusted components).
  • Trading of dependability and adaptability with other non-functional requirements like integrity (consistency) or performance. Approaches to improve the scalability of dependable and adaptive systems.
  • Foundations and formal methods (e.g. rigorous development of dependable systems, verification and refinement of fault tolerant systems, techniques and mechanisms ensuring application level fault tolerance).
  • System design, modeling, development and tool support for dependable and adaptive systems:
    • Software engineering methods and design patterns, including UML- and MDA-support for dependability and adaptiveness.
    • Tool support for coordination of complex services.
    • Design and programming abstractions to manage the complexity of adaptive mechanisms.
    • Policies and decision making for adaptation.
    • Aspect-oriented programming for dependability.
    • Metrics.
  • Evaluation and experience reports of dependable and adaptive distributed systems and services:
    • Experience reports from various application areas such as context-aware pervasive computing, safety-critical CNS systems (communication, navigation, and surveillance), embedded systems (e.g.automotive), and adaptive technologies in mobile computing.
    • Comparison of detection-, containment-, recovery-, and fault tolerance-strategies for distributed systems.
    • Application of middleware techniques to support reconfigurability and/or adaptability.

 

Track Program Chairs

Karl M. Göschka
Vienna University of Technology
Institute of Information Systems
Distributed Systems Group
Argentinierstrasse 8/184-1
A-1040 Vienna, Austria
phone: +43 664 180 6946
fax: +43 664 188 6275
Karl dot Goeschka (at) tuwien dot ac dot at

Svein O. Hallsteinsen
SINTEF ICT
Software Engineering Department
Andersens vei 15 b
NO-7465 Trondheim, Norway
phone: +47 7359 3010
fax: +47 7359 3350
Svein dot Hallsteinsen (at) sintef dot no

Rui Oliveira
Universidade do Minho
Computer Science Department
Campus de Gualtar
4710-057 Braga, Portugal
phone: +351 253 604 452 / Internal: 4452
fax: +351 253 604 471
rco (at) di dot uminho dot pt

Alexander Romanovsky
University of Newcastle upon Tyne
School of Computing Science
Office: Room 1008 , Claremont Tower
Newcastle upon Tyne, NE1 7RU, United Kingdom
phone: +44-191-222- 8135
fax: +44-191-222- 8788
Alexander dot Romanovsky (at) newcastle dot ac dot uk

Organisational Chair

Johannes Osrael
Vienna University of Technology
Institute of Information Systems
Distributed Systems Group
Argentinierstrasse 8/184-1
A-1040 Vienna, Austria
phone: +43 1 58801 58409
fax: +43 1 58801 18491


Program Committee

  • Michael Butler, University of Southampton (UK).
  • Emmanuel Cecchet, Emic Networks and ObjectWeb open source consortium (France).
  • Vicent Cholvi, Universitat Jaume (Spain)
  • Schahram Dustdar, Vienna University of Technology (Austria)
  • Frank Eliassen, University of Oslo (Norway).
  • Pascal Felber, Université de Neuchâtel
  • Jacqueline Floch, Stiftelsen for industriell og teknisk forskning - SINTEF (Norway).
  • Pablo Galdámez-Saiz, Universidad Politécnica de Valencia (Spain).
  • Kurt Geihs, University of Kassel (Germany). 
  • Matti Hiltunen, AT&T Labs (USA)
  • Istvan Majzik, Budapest University of Technology and Economics (Hungary)
  • Veena B. Mendiratta, Lucent Bell Labs (USA)
  • Francisco D. Muñoz-Escoí, Universidad Politécnica de Valencia (Spain).
  • Simin Nadjm-Tehrani, Linköping University (Sweden).
  • Marta Patino-Martinez, Universidad Politecnica de Madrid (Spain)
  • Jose Pereira, Universidade do Minho (Portugal)
  • Fernando Pedone, Università della Svizzera Italiana (Switzerland).
  • Luís Rodrigues, University of Lisbon (Portugal).
  • Andre Schiper, Ecole Polytechnique Federale de Lausanne (Switzerland)
  • Kaisa Sere, Åbo Akademi University (Finland).
  • Stefan Tai, IBM Watson (USA)
  • Steve Vinoski, IONA (USA)
  • Matthias Wiesmann, JAIST (Japan)
  • Mario Zenha-Rela, University of Coimbra (Portugal)
 

Final Program

The DADS track takes place on Thursday, 27th of April, 2006.

8h30 - 10h00:

The first session focuses on general and architectural mechanisms for adaption and fault tolerance:

1. Verification of Coordinated Exception Handling:
  (Fernando Castor Filho, Alexander Romanovsky, Cecilia Mary F. Rubira)
  In this paper, the authors present an approach to modeling and verifying fault-tolerant distributed systems that use exception handling as the main fault tolerance mechanism.
   
2. Combining Supervised and Unsupervised Monitoring for Fault Detection in Distributed Computing Systems:
  (Haifeng Chen, Guofei Jiang, Cristian, Ungureanu, Kenji Yoshihira)
  First a new concept of supervised and unsupervised monitoring is proposed for system fault detection. Second, in order to test the performance of the proposed approach, the authors simulate a variety of system faults in a real e-commerce application based on the multi-tiered J2EE architecture.
   
3. Modeling of Component-Based Adaptive Distributed Applications:
  (Kurt Geihs, Mohammad Ullah Khan, Roland Reichle, Arnor Solberg, Svein Hallsteinsen, Simon Merral)
  In this paper, the authors present an approach for modeling adaptation of component based distributed applications. The approach supports component-based design of different variants of the applications, and a framework for selecting proper variants based on the current state of the execution environment and the user preferences. XML is used as the specification language.
   
4. Proactive Resilience through Architectural Hybridization:
  (Paulo Sousa, Nuno Ferreira Neves, Paulo Verissimo)
  In this paper, the authors present the Proactive Resilience Model (PRM) and describe a design methodology under the PRM. This methodology is a way of building systems which guaranteedly do not suffer more than the assumed number of faults, and it is used to derive a distributed intrusion-tolerant secret sharing system.

 

10h30 - 12h00:

The second session concentrates on protocols assisting and enabling dependable and adaptive systems:

5. Service Interface: A New Abstraction for Implementing and Composing Protocols:
  (Olivier Rutti, Pawel T. Wojciechowski, Andre Schiper)
  In this paper the authors compare two approaches to the design of protocol frameworks – tools for implementing modular network protocols. The most common approach uses events as the main abstraction for a local interaction between protocol modules. The authors argue that an alternative approach, that is based on service abstraction, is more suitable for expressing modular protocols.
   
6. Post-Partition Reconciliation Protocols for Maintaining Consistency:
  (Mikael Asplund, Simin Nadjm-Tehrani)
  In this paper the reconciliation process is studied with a focus on integrity constraint conflicts and the decisions that affect the design of reconciliation algorithms. To compare different algorithms the authors have both defined a theoretical correctness criterion and experimental metrics. Three reconciliation algorithms have been introduced.
   
7. From Spontaneous Total Order to Uniform Total Order: different degrees of optimistic delivery:
  (Luis Rodrigues, Jose Mocito, Nuno Carvalho)
  This paper makes an overview of different optimistic total order protocols and shows how they can be combined in a single adaptive protocol.
   
8. Revisiting 1-Copy Equivalence in Clustered Databases:
  (Rui Oliveira, Jose Pereira, Afranio Correia Jr., Edward Archibald)
  In this paper the authors point out that 1-copy equivalence for causal consistency, which is subsumed by both serializability and snapshot isolation criteria, depends on basic session guarantees that are costly to ensure in clusters, especially in a multi-tier environment. Consequently, a simple solution is proposed, that guarantees causal consistency in the Database State Machine protocol and its performance is evaluated, thus highlighting the cost of seamlessly providing common consistency criteria of centralized databases in a clustered environment.

 

14h00 - 15h30:

The third session covers P2P and mobile/ad hoc systems:

   
9. Quiescent Consensus in Mobile Ad-hoc Networks using Eventually Storage-Free Broadcasts:
  (Francois Bonnet, Paul Ezhilchelvan, Einar Vollset)
  This paper presents a solution to solve the consensus problem using a new class of broadcasts that are appropriate to ad-hoc networking.
   
10. Evaluating Unstructured Peer-to-Peer Lookup Overlays:
  (Idit Keidar, Roie Melamed)
  In this paper, the authors define metrics for evaluating unstructured overlays for peer-to-peer lookup systems. These metrics capture the search dependability and efficiency, and the granularity at which one can control the trade-off between the two, as well as fairness.
   
11. An Adaptive Index Allocation Scheme for Reliable Data Retrieval and Provision in Peer-to-Peer Networks:
  (Atsushi Ito, Tomoyuki Ohta, Kouichi Mitsukawa, Yoshiaki Kakuda)
  This paper proposes an adaptive scheme for data retrieval and provision in peer-to-peer networks, in which indices are dynamically allocated in appropriate nodes adaptively to variation of traffic patterns caused by query messages.