Torsten Braun

Implementation of an Internet video conferencing application over ATM

The paper describes a complete implementation of an Internet video conferencing application taking advantage of the quality-of-service (QoS) capabilities provided by ATM. RSVP has been used to exchange the QoS parameters among the communicating video conferencing applications. The RSVP QoS parameters are mapped to ATM parameters and dedicated VCs with the corresponding ATM parameters are established by the underlying classical IP over ATM implementation. These VCs provide the desired QoS from the ATM network to the application data flow. The implementation is based on existing components such as an Internet conferencing tool, a RSVP implementation, and a classical IP over ATM implementation. The paper presents the required modifications of all these components to support QoS communication and the adaptation for their integration.

Multicast Communication In Large Scale Networks

Upcoming applications, such as computer supported cooperative work, distributed parallel processing and virtual shared memory depend on efJicient data exchange among multiple computer systems. They often require a reliable multipoint conzniunication service with low end-to-end delay and high throughput. Additional problem occur for the provision of a reliable multipoint service, where errors are more likely and where the sender has to deal with n m y receivers. i n order to meet the required reliability, and in order to allow an eficient use of network resources, powe@l error control mechanisms are essential. This paper presents a novel concept to support reliable multicast conimuitication in large scale heterogeneous networks. correction schemes suitable to provide reliable multicast service on top of unreliable network service. Common protocols use Go-Back-N or selective repeat to retransmit lost and corrupted data. Receivers request missing data directly from the multicast sender without consideration of network topology and actuaI network load. In case of multicast, it may be preferable to contact another communication participant placed closer to the requesting endsystem. This paper presents a novel error control mechanism suitable for global heterogeneous networks. It aims to reduce transfer delay, network load and the acknowledgement processing overhead at the sending side. Section 2 presents the proposed framework based on the concept of local groups. Section 3 gives results of a performance analysis with a simple example scenario. Finally, section 4 presents concluding remarks.

A parallel implementation of XTP on transputers

High performance communication systems have to provide very high data rates of more than 100 Mbit/s to new applications such as multimedia applications. To overcome the bottleneck in communication systems, the protocol processing of the layers above media access control layer, light weight protocols such as XTP (Xpress Transfer Protocol) are used. XTP is a protocol with functionality of layers three and four and was designed for protocol processing in reliable high speed networks such as FDDI, XTP is based on parallel finite state machines (FSMs) to support an efficient realization in VLSI. This paper presents a software implementation of XTP for a multiprocessor system. The implementation is based on these FSMs, which are mapped onto processes running on the processors of a multiprocessor architecture and communicating with each other by message passing.<<ETX>>

Internet protocols for multimedia communications. I. IPng-the foundation of Internet protocols

Surveys recent and current standardization work by the Internet Engineering Task Force (IETF) on Internet communication protocols. The author describes the IPng project and the IPv6 protocol that will form the basis for upper-layer protocols such as resource reservation, transport and application level protocols. The IETF decided in 1992 to develop a new version of the Internet Protocol (IP) because the available address space of IP version 4 (IPv4) will probably run out early in the 21st Century. The need for a new IP version provided the opportunity to redesign the protocol and add new functions.

Quality-of-Service Support for IP Flows over ATM

This paper describes an extension of a classical IP over ATM implementation which allows to support quality-of-service for IP flows. While standard classical IP over ATM implementations use shared VCs for several IP flows between two end systems, the classical IP over ATM implementation extensions allow to establish separate VCs with individual quality-of-service parameters for single IP flows. The paper describes the modifications of an AIX 4.2 kernel implementation. Mbone video applications using RSVP for resource reservation have been implemented on top of the classical IP over ATM implementation.

Dynamic bandwidth allocation for stored VBR video in ATM end systems

The paper introduces a scheme for delivery of variable bit-rate (VBR) coded video from a video server over ATM networks. To solve the problem of long term variations in bandwidth requirements, a dynamic bandwidth allocation scheme is proposed. This is based on renegotiation of bandwidth during the duration of the call.

Automated code generation for integrated layer processing

ALF (application level framing) and ILP (integrated layer processing) are protocol design and implementation concepts applied in high-performance communication architectures, e.g. to support multimedia applications. Writing ILP code is rather complex and, therefore, ILP code generation tools can reduce the time to develop efficient ILP protocol code significantly. This paper presents a tool, which allows automated ILP code generation based on high-level specifications of user data types. The tool is based on a stub compiler and integrates ILP loops into (un)marshalling routines automatically.

ALFred, a protocol compiler for the automated implementation of distributed applications

This paper describes the design and the prototyping of a compiling tool for the automated implementation of distributed applications: ALFred. This compiler starts from the formal specification of an application written in ESTEREL and then integrates end-to-end communication functions tailored to the application characteristics (described in the specification); it finally produces a high performance implementation. The paper describes the communication architecture associated with the approach. The compiler consists of a control compiler, also called ALF compiler, and a data manipulation compiler (the ILP compiler) that combines data manipulation functions in an efficient way (the ILP loop). The ALFred compiler has been designed to allow the development and the analysis of non-layered high performance communication architectures based on ALF and ILP.

Limitations and Implementation Experiences of Integrated Layer Processing

Integrated Layer Processing (ILP) has several limitations on current system environments. Performance evaluations show that the main benefits from ILP are rather a reduced number of memory accesses than an improved cache hit rate. Data manipulation characteristics have significant impact on the cache behavior and on the achievable performance gain of ILP. Future protocol characteristics and efficient implementation concepts for communication systems with CPU expensive data manipulations are proposed.

On Transport Systems for ATM Networks

Emerging ATM networks accompanied by an increasing variety of application service requirements may change future transport subsystem design in order to provide enhanced service characteristics and improved performance. System designs integrating proper architectures, services and protocols are increasingly required. This paper gives a survey of some research work dealing with transport subsystem issues for ATM network. A framework for an ATM transport system is presented providing an enhanced service interface and supporting application-driven flexible protocol configurations. The framework uses a non-hierarchical, fine granular function-based decomposition of the communication task in order to efficiently provide these features.