作者：
Tutschku, KurtMüller, PaulTran, Freádeáric DangFaculty of Computer Science
Institute of Distributed and Multimedia Systems University of Vienna Universitaetsstrasse 10/T11 A-1090 Vienna Austria
Faculty of Communication Sciences Technical University of Kaiserslautern Paul-Ehrlich-Strasse Bld. 34 D-67653 Kaiserslautern Germany Orange Labs
France Teáleácom 38-40 rue du Geáneáral Leclerc F-92794 Issy-les-Moulineaux Cedex 09 France
Vehicular Ad-hoc Networks are a very interesting self-organizing communication system that can make the traffic on our roads much safer. A lot of aspects of the behavior of these systems are studied using network simu...
详细信息
ISBN:
(纸本)9781565553415
Vehicular Ad-hoc Networks are a very interesting self-organizing communication system that can make the traffic on our roads much safer. A lot of aspects of the behavior of these systems are studied using network simulation frameworks. While the upper communication layers of Vehicular Ad-hoc Networks are mainly simulated according to their specifications, in this paper we will highlight weak points in these simulations. They occur due the fact that the layers below the link layer are considered very rarely and are also almost neglected by the commonly used network simulation frameworks. This can potentially lead to inaccurate simulation results which can be fatal when studying safety relevant applications or application with real-time constraints. In this work in progress paper we are going to propose an approach to get this under control.
The European STREP project Quasimodo1 develops theory, techniques and tool components for handling quantitative constraints in model-driven development of real-time embeddedsystems, covering in particular real-time, ...
详细信息
ISBN:
(纸本)9781605589046
The European STREP project Quasimodo1 develops theory, techniques and tool components for handling quantitative constraints in model-driven development of real-time embeddedsystems, covering in particular real-time, hybrid and stochastic aspects. This tutorial highlights the advances made, focussing on real industrial case studies tackled.
In service-oriented systems, clients can access services via a network. Service level agreements (SLA) can exist, which specify - among other things - performance-related Quality of Service (QoS) properties between th...
详细信息
ISBN:
(纸本)9781450301077
In service-oriented systems, clients can access services via a network. Service level agreements (SLA) can exist, which specify - among other things - performance-related Quality of Service (QoS) properties between the client and the server, such as round-trip time, processing time, or availability. For a service provider serious financial consequences or other penalties can follow in case of not fulfilling the SLAs. The service consumer wants to evaluate that the provider complies with the guaranteed SLAs. Designing and developing a QoS-aware service-oriented system means facing many design challenges, such as where and how to measure the performance-related QoS properties. This paper presents design practices and patterns for measuring such QoS properties by extending and utilizing existing patterns. The focus of the patterns lies on the QoS measuring impact on the client's or service's performance, the extend of separation of concerns, the property of reusability, and the preciseness of the measured QoS properties. The patterns help to build efficient solutions to measure performance-related QoS properties in a service-oriented system.
In this paper, we study the theory of collaborative upload bandwidth measurement in peer-to-peer environments. A host can use a bandwidth estimation probe to determine the bandwidth between itself and any other host i...
详细信息
In this work, we study the target detection and tracking problem in mobile sensor networks, where the performance metrics of interest are probability of detection and tracking coverage, when the target can be stationa...
详细信息
Web services are going mobile. A Mobile Enterprise can be established in a cellular network by participating Mobile Hosts, which act as web service providers, and their clients. Mobile Hosts enable seamless integratio...
详细信息
Mashup platforms and end-user centric composition tools have become increasingly popular. Most tools provide Web interfaces and visual programming languages to create compositions. Much of the previous work has not co...
详细信息
ISBN:
(纸本)9781450304184
Mashup platforms and end-user centric composition tools have become increasingly popular. Most tools provide Web interfaces and visual programming languages to create compositions. Much of the previous work has not considered compositions comprising human provided services (HPS) and software-based services (SBS). We introduce a novel HPS aware service mashup model which we call socially oriented mashups (SOM). The inclusion of HPS in service mashups raises many challenges such as a QoS model that must account for human aspects and the need for flexible execution of mashups. We propose human quality attributes, for example delegation, and a context model capturing various information including location and time. The QoS and context model is used at design-time and for runtime adaptation of mashups. In this paper, we show how to model context-aware SOMs that include HPS and SBS and demonstrate the first results of our working prototype. Copyright 2010 ACM.
In this paper we present the implementation of a framework for accelerating training and classification of arbitrary Convolutional Neural Networks (CNNs) on the GPU. CNNs are a derivative of standard Multilayer Percep...
详细信息
In this paper we present the implementation of a framework for accelerating training and classification of arbitrary Convolutional Neural Networks (CNNs) on the GPU. CNNs are a derivative of standard Multilayer Perceptron (MLP) neural networks optimized for two-dimensional pattern recognition problems such as Optical Character Recognition (OCR) or face detection. We describe the basic parts of a CNN and demonstrate the performance and scalability improvement that can be achieved by shifting the computation-intensive tasks of a CNN to the GPU. Depending on the network topology training and classification on the GPU performs 2 to 24 times faster than on the CPU. Furthermore, the GPU version scales much better than the CPU implementation with respect to the network size.
暂无评论