Dynamic Resource Allocation for Multimedia Document Retrieval over High Speed LANs

The emergence of gigabit local area networks (G-LANs) has spurred a tremendous interest in supporting networked multimedia applications over a LAN. In this paper, we propose a mechanism for dynamically allocating network resources in asynchronous LANs. Presentation of multimedia objects with required play-out quality requires Quality of Service (QoS) guarantees by the underlying networking infrastructure. Existing asynchronous LANs, such as Ethernet, do not support the notion of QoS due to their asynchronous media access protocol. For such networks, we propose a dynamic bandwidth management scheme that uses the concept of Time Division Multiple Access (TDMA). Significant performance improvement is observed through experimental results. In particular, the transmission rates for multimedia hosts improve significantly with low jitter variations in media streams. We also propose a framework for graceful degradation of play-out quality of multimedia objects in case the LAN's total capacity is not sufficient to meet the overall demand.     

Are ATM, Gigabit Ethernet ready for prime time

ATM and Gigabit Ethernet, the subjects of intense speculation by industry observers and potential users of the technologies, appear as if they may be ready to begin competing for their place in the networking sun. The demand for both technologies is being caused by the increasing adoption of Internet and intranet technology and the growing number of bandwidth-hungry applications. These factors make it necessary to move an increasing amount of data at faster rates over networks. Enterprises are thus beginning to adopt ATM and Gigabit Ethernet. In some cases, the two technologies are replacing FDDI backbones because they can be less expensive and provide higher speeds. In addition, ATM has quality of service features, while Gigabit Ethernet can provide seamless connectivity throughout LANs that already use Ethernet technology. Both technologies provide LAN services, such as scalable campus backbones, load sharing, and connections to WAN services. ATM has been available longer and is being deployed where high bandwidth performance is required. ATM and Gigabit Ethernet involve a variety of complex issues that will, in part, determine whether and how users adopt either or both technologies     

Asynchronous Transfer Mode and other Network Technologies for Wide-Area and High-Performance Cluster Computing

We review fast networking technologies for both wide-area and high performance cluster computer systems. We describe our experiences in constructing asynchronous transfer mode (ATM)-based local- and wide-area clusters and the tools and technologies this experience led us to develop. We discuss our experiences using Internet Protocol on such systems as well as native ATM protocols and the problems facing wide-area integration of cluster systems. We are presently constructing Beowulf-class computer clusters using a mix of Fast Ethernet and Gigabit Ethernet technology and we anticipate how such systems will integrate into a new local-area Gigabit Ethernet network and what technologies will be used for connecting shared HPC resources across wide-areas. High latencies on wide-area cluster systems led us to develop a metacomputing problem-solving environment known as distributed information systems control world (DISCWorld). We summarize our main developments in this project as well as the key features and research directions for software to exploit computational services running on fast networked cluster systems.     

An experimental analysis of DCT-based approaches for fine-grained multiresolution video

In the future, video-streaming systems will have to support adaptation over an extremely large range of display requirements (e.g., 90×60 to 1920×1080). This paper presents the architectural trade-offs of bandwidth efficiency, computational cost, and storage cost to support fine-grained multiresolution video over a large set of resolutions. While several techniques have been proposed, they have focused mainly on limited spatial resolution adaptation. In this paper, we examine the ability of current techniques to support wide-range spatial resolution adaptation. Based upon experiments with real video, we propose an architecture that can support wide-range adaptation efficiently. Our results indicate that multiple encodings with limited spatial adaptation from each encoding provide good trade-offs between efficient coding and the ability to adapt the stream to various resolutions. Jie Huang received her BS in computer and communications and MS in computer science from Beijing University of Posts and Telecommunications, Beijing, China, in 1992 and 1995 respectively, where she was an assistant professor from 1995 to 1999. Since 1999, she has been pursuing her PhD at OGI school of Science and Engineering at Oregon Health and Science University (from 1999 to 2004) and Portland State University (since 2004). Her research interests include multimedia networking and software engineering. Wu-chi Feng received his Ph.D. in Computer Science and Engineering from the University of Michigan in 1996. ~His research interests include multimedia systems, video-based sensor networking technologies, and networking. ~He currently serves as an Editor for the Springer-ACM Multimedia Systems Journal. ~He also serves on the national Orion Cyberinfrastructure Advisory committee. Jonathan Walpole received his Ph.D. degree in Computer Science from Lancaster University, UK. He is a Professor in the Computer Science Department at Portland State University. Prior to joining PSU he was a Professor and Director of the Systems Software Laboratory at the OGI School of Science and Engineering at Oregon Health & Science University. His research interests are in operating systems, networking, distributed systems and multimedia computing. He has pioneered research in adaptive resource management and the integration of application and system-level quality of service management. He has also done leading edge research on dynamic specialization for enhanced performance, survivability and evolvability of large software systems. His research on distributed multimedia systems began in 1988, and in the early 1990s he lead the development of one of the first QoS-adaptive Internet streaming video players.     

QoS-Aware Admission Control and Dynamic Resource Provisioning Framework in Ubiquitous Multimedia Computing Environments

A ubiquitous service deployment is emerging in the multimedia, networking, and wireless mobile computing area. Therefore, there has been an increasing demand for ubiquitous computing environments to support a certain degree of quality of service (QoS) to meet various service requirements from different computing and networking applications, and to better utilize the computing resources. However, supporting QoS in the ubiquitous computing environments has also raised great concerns regarding the applicability of any QoS solution. Management of such ubiquitous multimedia applications requires new mechanisms, i.e., Soft-QoS framework, to be developed for admission control, negotiation, allocation, and scheduling. In this paper, we present a novel negotiated admission control algorithm that exploits the degradability property of applications to improve the performance of the system. The algorithm is based on setting aside a portion of the resources as reserves and managing it intelligently, so that the total utility of the system can be maximized. The mixed greedy and predictive strategy leads to an efficient protocol that also improves the system performance. We use the constructs of application benefit functions and resource demand functions in the integrated admission control and negotiation protocol. We applied our Soft-QoS framework to the admission controlling and resource scheduling for ubiquitous multimedia devices such as Continuous Media (CM) or Video-On-Demand (VOD) servers, where multimedia applications can generally tolerate certain variations on QoS parameters by providing multiple classes with consistently proportional rather than absolute QoS. Extensive simulation experiments are presented in the paper to evaluate the performance of the novel mechanisms and compare it against some other methods used in the past.This work was in part funded by DARPA through the SPAWARSYSCEN under Contract Number N66001-97-C-8525 and SK Telecom, Korea, under Contract Number (KU-R0405721).     

Code optimizations for a VLIW-style network processing unit

The explosive growth in network bandwidth and Internet services such as QoS (quality of service) and SLA (service level agreement) monitoring have created the need for new networking hardware called a Network Processing Unit (NPU). In order to rapidly reconfigure the NPU for frequently varying Internet services and technologies, a high-performance C compiler is urgently needed. Several code generation techniques, which are intended to meet the high code quality demands of other types of application specific instruction-set processors (ASIPs) like digital signal processors (DSPs), have already been developed. However, these techniques are insufficient for NPUs due to striking architectural differences such as asymmetric data paths. The main purpose of this paper is to discuss our recent experience with the development of a commercial compiler for a new NPU called the Paion PPII, which is basically a packet engine for NPU to meet the growing need for new high-bandwidth communication equipment targeted for Internet routers and ethernet adapters. For this purpose, we will first show the architectural challenges posed by the target NPU. Then, we will describe several compiler techniques that we found to be effective for the target NPU with various unorthogonal architectural features. The current implementations of the PPII use a VLIW (Very Long Instruction Word) architecture. So, we handled this VLIW-style architecture by employing a simple code compaction scheme which packs multiple parallel instructions into one long instruction word. The experimental results show that our techniques are effective for significantly reducing the dynamic instruction count. Copyright © 2004 John Wiley & Sons, Ltd.     

Computer communications TC

To fully realise the promise of emerging high speed networks we need: high bandwidth applications; networking standards; and networking services based on those standards. Two service models are being pursued to support high speed multimedia applications: IETF (Internet Engineering Task Force) and ATM (asynchronous transfer mode). The authors compare these models. They then discusses the future activities of the Computer Communications Technical Committee in this regard     

支持Internet实时多媒体应用的Overlay技术研究

目前出现的大规模（large-scale）的多媒体应用对Internet提出了新的挑战。传统的IP组播（IP multicast，IPM）由于实现技术的复杂性，很难在Internet上推广。相反，基于叠加网络（overlay network，ON）的解决方案则受到各方面的关注。本文针对大规模实时多媒体应用需求，通过建立评价参数模型对ON协议进行研究。尤其时overlay组播（overlay multicast，OM）的QoS控制、可靠性以及可伸缩性机制进行了分析。与IPM相比，OM虽然有性能代价（performance penalty）问题，但是OM能方便、灵活地利用多种高层的功能实现，基于hop-by-hop方式有效地提供E2E（end—to-end）的可靠性和QoS保障，并且较好地支持TCP-friendliness，具有良好的可伸缩性。此外，0M的基础网络（infrastructure／underlay network）无关性，也使其容易在Internet中进行配置。以上特性使得OM成为支持Internet上大规模实时多媒体应用的最具前景的技术．     

Evolving technologies for new Internet applications

What will drive the Internet as new multimedia applications demand an infrastructure that enables greater network capacity and support for multicast and quality of service than the traditional TCP/IP and router-based structure? This article surveys seven key technologies that show promise     

Group Decision Making on Urban Planning Using Desktop Multimedia Conferencing

Recent developments on multimedia systems and networking technology show that using desktop multimedia conferencing for group decision making on wide area networks such as the Internet is possible. In this paper we review the design, hardware and software requirements and organizational issues in a desktop multimedia conferencing system. We draw on our experiences from multiple multimedia conferences on the Internet and in particular we focus on a case study on urban planning using desktop multimedia conferencing on the Internet. Further we discuss implications for further research on desktop multimedia conferencing.     

QoS Management in Educational Digital Library Environments

Advances in multimedia computing technologies offer new approaches to the support of computer-assisted education and training within many application domains. Novel interactive presentation tools can be built to enhance traditional teaching methods with more active learning. Since a variety of user expectations are possible in such an environment, research must address the incorporation of these factors into presentation tools. During an interactive learning/training process, presentation tools must be able to handle various types of delays. A flexibly adjustable quality of service (QoS) should thus be supported. In this paper, we investigate a framework and systematic strategies for supporting the continuous and synchronized retrieval and presentation of multimedia data streams in a client/server distributed multimedia environment for educational digital libraries. Specifically, we establish a practical framework for specifying multimedia objects, tasks, schedules, and synchronization constraints between media streams. We identify the QoS parameters critical to the support of multimedia presentations for learning and training activities. Based on the proposed framework and QoS specifications, we develop presentation scheduling and buffer management strategies which can enforce the specified QoS requirements in an educational digital library environment.     

IMS Application Servers: Roles, Requirements, and Implementation Technologies

The IP multimedia subsystem (IMS) defines a generic architecture to support communication services over a Session Initiation Protocol (SIP) infrastructure. In the IMS architecture, application servers host and execute the IMS service logic. These servers can be SIP application servers, open services architecture (OSA) application servers, or a customized applications for mobile networks using enhanced logic (Camel) service environment. Some technologies used in telephony and voice-over-IP (VoIP) application servers are also applicable to IMS application servers, but such servers have some unique requirements that could limit the extent to which these technologies can meet them.     

An open framework supporting multimedia web services

With the rapid emergence of Web services, more and more Web services are published on the Internet as resources for Web application development. There may exist some relationships among different Web services, such as exact match, plug-in match, and irrelevant. In this paper, we discuss a set of requirements related to multimedia Web services, and propose a three-tier framework to establish an open environment supporting multimedia Web services, while partially implementing the requirements. This paper focuses on the design of the service broker tier that is essential for future Web services-oriented system design and integration and enabling Web services more transparent, interoperable, and fault-tolerate.     

Low-cost,delay-bounded point-to-multipoint communication to support multicasting over WDM networks

The confluence of technical advances and multimedia service needs is intensifying the need for high throughput and low latency. Future communication networks will face an increase in traffic driven by multimedia requirements with stringent delay and jitter requirements. Wavelength division multiplexing (WDM) optical networks have the potential for meeting these goals by offering unprecedented high bandwidth and low latency. One very important aspect of the emerging Internet services is the need to support multicasting. This is crucial if WDM networks were to play an efficient role in the next generation Internet.Multicasting in WDM networks supporting multimedia applications can be viewed as the process of taking a group communication request and selecting a multicast tree that satisfies the quality of service requirements, in terms of bandwidth and end-to-end delay, of the underlying application. In this paper, we present a new class of low-cost, bounded-delay multicast heuristics for WDM networks. The heuristics use various techniques to establish a tree of semi-lightpaths between a source and a group of destination nodes. The unique feature of these heuristics is that they decouple the cost of establishing the multicast tree from the delay incurred by data transmission due to light-wave conversion and processing at intermediate nodes along the transmission path. A simulation study shows the performance of the proposed heuristics.     

Distributed computing architecture for effective management of multimedia streams on DPE

We study a networking architecture model that is built on a distributed processing environment (DPE) for multimedia services suitable for high speed transport networks such as ATM networks. In this architecture, the applications are deployed as units of software building blocks. Each building block provides a layered view for the effective management and control of the multimedia network resources and services according to the concept of telecommunications management network (TMN) and telecommunications information networking architecture (TINA). For the purpose of flexible service provision to users and effective service introduction by service providers, this architecture proposes the adoption of ad hoc service building blocks such as a video on demand building block and a CSCW building block that have interactions with a general purpose building block. This paper also proposes a naming structure for the management of user profiles and session profiles using a directory service system, and an effective control model for multimedia logical device objects using a stream process approach. The proposed model is implemented on a DPE platform that provides various transparencies, ANSAware.     

Editorial: Internetware in Motion

The Internet provides a global open infrastructure for exchanging and sharing of various resource all over the world. The rapid development and wide application of the Internet makes it a new mainstream platform to use, develop, deploy and execute software systems and applications. With the vision of "Internet as a computer", many technical initiatives such as pervasive computing, grid computing, service computing and cloud computing emerges on this open and dynamic environment. In order to support the various new application styles and accommodate the fundamental change of the underlying infrastructure, many specific software technologies such as service-oriented architecture are proposed for current practices. While these technologies are useful and widely accepted, they have not formed a systematic solution as matured as the object-oriented technology, as a uniformed software methodology and technology system is yet to be developed.     

Internet traffic measurement

The Internet's evolution over the past 30 years (1971-2001), has been accompanied by the development of various network applications. These applications range from early text-based utilities such as file transfer and remote login to the more recent advent of the Web, electronic commerce, and multimedia streaming. For most users, the Internet is simply a connection to these applications. They are shielded from the details of how the Internet works, through the-information-hiding principles of the Internet protocol stack, which dictates how user-level data is transformed into network packets for transport across the network and put back together for delivery at the receiving application. For many networking researchers however, the protocols themselves are of interest. Using specialized network measurement hardware or software, these researchers collect information about network packet transmissions. With detailed packet-level measurements and some knowledge of the IP stack, they can use reverse engineering to gather significant information about both the application structure and user behavior, which can be applied to a variety of tasks like network troubleshooting, protocol debugging, workload characterization, and performance evaluation and improvement. Traffic measurement technologies have scaled up to provide insight into fundamental behavior properties of the Internet, its protocols, and its users. The author introduces the tools and methods for measuring Internet traffic and offers highlights from research results     

Recent advances in multimedia networking

In recent years, the ubiquity of multimedia services along with the proliferation of mobile devices and the demand for new audio and video applications are changing the life style of users. User demands for multimedia access anywhere, anytime from any device are creating new challenges for research communities from both academia and industry. It is expected that video-based services alone will account for 50 percent of all consumer network traffic in 2012 and we will continue to witness the explosive growth in users sharing multimedia content over the Internet. In this context, new network, application, and user-based approaches must be created to deal with such complex multimedia systems. This paper presents some of the recent advances in multimedia networking focusing primarily on areas that have been receiving attention recently and are expected to continue to generate further interests in coming years. These areas include Quality of Experience (QoE) and various related standardization issues, Content Distribution Networks (CDNs), multimedia communications, mobile Multimedia. This paper also briefly highlights some of the major challenges that still need to be addressed to enable the support and delivery of multimedia services anywhere, anytime over highly heterogeneous infrastructures and user terminal devices.     

流媒体对象的缓存管理策略

基于流媒体服务的代理技术是流媒体研究领域中的重要课题.随着流媒体技术在Internet和无线网络环境中的高速发展,对流媒体代理服务器的研究也正在逐步深入.本文主要讨论通过代理技术改善媒体的服务质量,降低媒体的传输延迟以及减轻网络负载.在Internet环境下,对流媒体代理服务器的研究集中于流媒体的访问特性、缓存替换算法,构建和实现一个流媒体代理服务器是对流媒体代理技术研究的基础.