The proceedings contain 27 papers. The special focus in this conference is on Modeling, Resource Allocation, Admission Control, Multimedia and Incentives. The topics include: Modelling, measurements, and admission con...
ISBN:
(纸本)3540402810
The proceedings contain 27 papers. The special focus in this conference is on Modeling, Resource Allocation, Admission Control, Multimedia and Incentives. The topics include: Modelling, measurements, and admission control;statistical characterization for per-hop QOS;performance analysis of server sharing collectives for content distribution;an approximation of the end-to-end delay distribution;price-based resource allocation in wireless ad hoc networks;on achieving fairness in the joint allocation of processing and bandwidth resources;distributed admission control for heterogeneous multicast with bandwidth guarantees;subjective impression of variations in layer encoded videos;a moving average predictor for playout delay control in VOIP;a game-based control-theoretic approach to peer-to-peer incentive engineering;improving dependability of real-time communication with preplanned backup routes and spare resource pool;fault tolerance in networks with an advance reservation service;routing and grooming in two-tier survivable optical mesh networks;fast network re-optimization schemes for MPLS and optical networks;hotspot mitigation protocol for mobile ad hoc networks;failure insensitive routing for ensuring service availability;network availability based service differentiation;replica placement for widely distributedsystems;using latency quantiles to engineer QOS guarantees for web services;dynamic resource allocation for shared data centers using online measurements;providing deterministic end-to-end fairness guarantees in core-stateless networks;per-domain packet scale rate guarantee for expedited forwarding;online response time optimization of apache web server and a practical learning-based approach for dynamic storage bandwidth allocation.
Adaptive resource management (ARM) technology has emerged as a way to: (1) support applications' varying resource needs, and (2) provide reusable solutions to commonly occurring problems. Most existent works on AR...
详细信息
Adaptive resource management (ARM) technology has emerged as a way to: (1) support applications' varying resource needs, and (2) provide reusable solutions to commonly occurring problems. Most existent works on ARM have focused on centralized approaches. However, centralized ARMs are not scalable, which may result in slow responses and wrong allocation decisions based on inaccurate system state information. In this paper, we show how to address these problems with a decentralized resource management approach. Based on social models, the approach appears to have the potential to provide prompt and accurate allocation of resources.
Current models of component architectures require extensions to support compositional reasoning about extra-functional properties such as worst-case time. Studying such properties in architectures is complicated since...
详细信息
Current models of component architectures require extensions to support compositional reasoning about extra-functional properties such as worst-case time. Studying such properties in architectures is complicated since actual components may not yet be chosen during architectural design, and different choices may have significant and hard to predict effects on the system properties. In this paper we show how finite state automata and Petri nets may be extended to provide compositionality of extra-functional properties. We focus on worst-case time and safety. We illustrate the use of these techniques on the well-known production cell case study. In collaboration with industry we are currently applying a prototype analysis system for predicting critical properties of real-time industrial control systems.
To safely establish the correct timing of a real-time processing node, adequate architectural structures have to be used. This refers to the hardware architecture of the processing node as well as the software archite...
详细信息
To safely establish the correct timing of a real-time processing node, adequate architectural structures have to be used. This refers to the hardware architecture of the processing node as well as the software architecture of its operating system and application software. This paper presents architectures that allow for a well structured and simple timing analysis. First, it presents solutions for cleanly splitting the overall timing analysis into schedulability analysis and task worst-case execution time analysis. Second, it presents a programming strategy that yields software that is highly temporally predictable and easy to analyze for its worst-case execution time.
作者：
L.R. WelchCenter for Intelligent
Distributed and Dependable Systems School of Electrical Engineering and Computer Science Ohio University Athens OH USA
Adaptive resource management middleware enables a pool of computing and network resources to be allocated at run-time. This allows systems to adapt to processing demands that depend on dynamic environments, and allows...
详细信息
Adaptive resource management middleware enables a pool of computing and network resources to be allocated at run-time. This allows systems to adapt to processing demands that depend on dynamic environments, and allows systems to reconfigure so that critical functions continue to be provided when some hardware components have failed. This paper presents a process that can be used on-line to find allocations that are feasible/schedulable (i.e., allocations that allow all realtime constraints to be met).
Efficient support for multi-class services is a very important issue in integrated wireless and mobile networks because each type of service has distinct characteristics and Quality of Service (QoS) requirements. In t...
详细信息
Efficient support for multi-class services is a very important issue in integrated wireless and mobile networks because each type of service has distinct characteristics and Quality of Service (QoS) requirements. In this paper, we propose and analyze a new channel allocation scheme for integrated wireless and mobile networks. In our scheme, traffic is grouped into four different classes: originating real-time service calls, originating non real-time service calls, real-time service handoff request calls, and non real-time service handoff request calls. This new scheme can improve the grade of service of higher priority classes while maintaining the required QoS for lower priority classes. An analytical model is developed based on a multidimensional Markov chain and validated by simulation results. The closed formula of several important system performance parameters, including blocking probability of. originating calls, forced termination probability of real-time service handoff request calls, and average transmission delay of non real-time service calls, are obtained.
In this paper we develop a methodology for treating the problem of scheduling partially-ordered tasks on parallel machines. Our framework is based on the timed automaton model, originally developed for verification of...
详细信息
In this paper we develop a methodology for treating the problem of scheduling partially-ordered tasks on parallel machines. Our framework is based on the timed automaton model, originally developed for verification of real-time programs and digital circuits and more recently adapted for solving time-optimal scheduling problems. In this framework, the scheduling problem admits a state-space representation and an optimal schedule corresponds to a shortest path in the timed automaton. We check our implementation on numerous benchmarks and show how release times and deadlines can be easily incorporated into the model.
The state-of-the-art intelligent vehicle, autonomous guided vehicle and mobile robotics application domains can be described as collection of interacting highly autonomous complex dynamical systems. Extensive formal a...
详细信息
The state-of-the-art intelligent vehicle, autonomous guided vehicle and mobile robotics application domains can be described as collection of interacting highly autonomous complex dynamical systems. Extensive formal analysis of these systems - except special cases - is not feasible, consequently the availability of proper simulation and test tools is of primary importance. This research targets the real-time hardware-in-the-loop (HIL) simulation of vehicle and mobile robot systems. To certain extent distributed virtual environment (DYE) systems are attempting to satisfy similar requirements but a few distinctive features set this approach apart. DVE systems put the emphasis on load balancing and communication overhead. In our case the emphasis is on the temporal predictability and guaranteed, timed execution of the experiment. The paper describes a simulation framework dedicated to HIL simulation of continuous dynamical entities with spatially constrained interactions. The underlying modelling concept is introduced. The runtime infrastructure is described, which allows for distributed execution of the models.
FORGE brings together a number of advances in architectural modeling, software architecture and distributed/real-timesystems to build a platform that provides two fundamental capabilities for distributed, realtime, ...
详细信息
FORGE brings together a number of advances in architectural modeling, software architecture and distributed/real-timesystems to build a platform that provides two fundamental capabilities for distributed, realtime, and embedded (DRE) system development: (a) conceptualization and coding of the design knowledge through collaborative specifications that are inherently matched to distributed solutions; and (b) exploitation of the design knowledge across all development phases for the DRE systems. Our proof-of-concept FORGE prototype is built upon collaborative specifications captured by extensions to the message sequence charts (MSCs) that drive the customization of CompOSEIQ middleware services and generate node-architecture specific code through descriptions of the architecture and resources captured using ADL and RDL respectively.
暂无评论