News

• Download of presentations is now available at the program page

Overview

The International Workshop on Parallel and Distributed Real-Time Systems is a forum for the presentation and discussion of approaches, research findings, and experiences in the domain of large-scale parallel and distributed real-time systems. Both research and development of relevant technologies are of interest, as well as the applications built using such technologies.

General Paper Sessions

These sessions will present high-quality papers submitted to the workshop and selected by the program committee for presentation and publication at WPDRTS.

Topics of interest include but are not limited to:

• Resource Management: Value-based, Feedback-based, and Power-aware scheduling, Combined scheduling of hard and soft real-time tasks, Dynamic real-time systems, Real-time servers, Fault-tolerance and Security in real-time systems.
• Operating Systems and Middleware: Run-time systems, Middleware architecures, Real-Time Linux, Real-Time CORBA, Real-time Databases.
• Programming Environments: Software design, Parallelization methods/tools for DSP-based, reconfigurable, and mixed-computation-paradigm architectures, Real-Time Java.
• Algorithms and Applications: Signal/image processing, Vision/robotic systems, Sensor Web, Industrial automation, Vehicle guidance, Command and control.
• Architectures: Special-purpose processors, mixed-computation-paradigm, size/weight/power modeling and management.
• Specification, Modeling, and Analysis: Formal methods, Object orientation, Benchmarking, Tools and environments.
• Networking and Communications: Real-time communication protocols, Sensor Networks, allocation control mechanisms, and performance analysis.

Special and Invited Sessions

Formal methods in distributed real-time systems

chairs: Angelika Mader, University of Twente, Netherlands and Ansgar Fehnker, University of New South Wales, Australia.

• The aim of this Special Session is to bring together researchers, designers and developers interested in supporting analysis and design of time critical applications by formal methods. The session is focused on: formal methods and tools for time critical applications (Timed Automata, Timed Petri Nets, Real-time testing,…), modelling approaches and analysis for time critical applications (e.g. Real-time UML), scheduling and optimization of systems with timing constraints, schedulability analysis of RT applications (RMA, Response Time Analysis,… ), case studies and other topics related to formalisms with deadline constraints.

Wireless sensor networks

chair: Tarek F. Abdelzaher, University of Illinois at Urbana-Champaign, USA.

• The purpose of this special session is to present and discuss the latest analytic, systems, and deployment challenges in wireless sensor networks. Such networks, featuring myriads of tiny devices equipped with sensing, local actuation, communication, and processing, offer significant new problems in the design of real-time communication protocols, middleware services, and programming abstractions for massively distributed wireless computing. They bring about a need for new models of computation and real-time performance analysis, as well as new theory on which such models are based. The session hopes to bring together experts, practitioners and researchers, from academia and industry, to present challanges and solutions in this growing field.

Feedback control and real-time systems

chair: Karl-Erik Arzen, Lund Institute of Technology, Sweden.

• The topics of the session include any issue related to control in real-time systems and the interaction between computing and control systems, in particular: computational models and languages for control applications; implementation-aware and resource-constrained control systems; integration of control and scheduling incl. feedback scheduling; co-design tools for control, computing and communication; modeling and simulation of performance control; temporally robust control systems; hybrid system approaches in integrated control and computing; applications of control to real-time computing

Soft and Firm Real-Time Systems (or Kinder, Gentler Real-Time)

chairs: Jeffe­ry Hansen, Car­negie Mellon University, USA and Frank Drews, Ohio University, USA.

• Rate-monotonic scheduling and other techniques for designing hard real-time systems are well established with many examples of deployed systems. Hard real-time systems generally assume that a single deadline miss results in failure of the system. For most real-world systems this assumption is overly pessimistic leading to significant waste of resources. Often allowing even very small fractions of deadlines to be missed (e.g, 0.00000001) can result in a significant reduction in resource requirements. An ideal scheduling algorithm would support the full spectrum of application real-time requirements from best effort to hard real-time. To be effective, soft/firm real-time systems must focus on more than just minimizing the mean delay. If delay variance is not kept under control, a reduction in mean delay may not significantly reduce (or even result in an increase of) the deadline miss rate. Good soft real-time scheduling algorithms will address the task delay distribution either with respect to a fixed deadline or through a time/utility function. Topics of interest include, but are not limited to: use of soft/firm real-time on traditionally hard real-time applications; real-time guarantees for dynamic and stochastic workload; probabilistic real-time guarantees; mixed soft/hard real-time systems and algorithms.