A goal of a broadband ISDN network is to provide integrated transport for a wide range of applications such as teleconferencing, entertainment video, and file distribution. These require multipoint communications in addition to conventional point-to-point connections. The essential component to provide multipoint communications is a multicast packet switch. In this paper, we propose and analyze a new parallel multicast packet switch which easily approaches a maximum throughput of 100% as the number of fanout and multicast rate are increased. The proposed switch consists of a simple ring network and a point-to-point switch network in parallel. The ring network provides both replication and routing of multicast packets. The point-to-point switch network is responsible for delivering only unicast packets. The ring network provided in this switch overcomes the problems of clock synchronization and unfairness of access in the slotted ring by synchronizing the ring to the time slot used in the point-to-point switch and providing small amount of speed-up. Moreover, the significant drawbacks of the basic cascaded multicast fabric design are removed in this parallel switch which can separate the unicast and multicast packets before entering the switch fabric. The performance analysis shows that this switch with the small size of input/output buffers achieves good performance in delay and throughput, and the packet loss probability less than 10-9.

A high-efficiency air cooling system is one of the keys to achieving high throughput in an ATM switching system for Broadband ISDN. Our approach is to cool the multichip modules plugged into a planar packaging system by using a two-phase thermosyphon cold-plate with an air-cooled condenser. Physically separating the cold-plate and the air-cooled condenser and connecting item by small diameter pipes is the key to applying this cooling technology to large planar packaging systems to increase volumetric packaging densities. Furthermore, thermosyphon technology allows the heat transfer process to operate without any external pumping power. Therefore this cooling system is regarded an extended high-performance air cooling system. The optimum structure was investigated while focusing on ways to reduce the external thermal resistance. The external thermal resistance between the system's cold-plate and air inlet was measured to be 0.21 K/W at an air velocity of 2 m/s and a cooling duty of 150 watts. Using this external thermal resistance value, we simulated the cooling characteristics of an MCM containing a 44 array of 10-mm-square LSI chips on an alumina substrate measuring 100100 mm. For an allowable temperature rise of 60, simulated thermal resistance was 6 K/W at an air flow of 2 m/s. This allows a power dissipation of more than 160 watts per MCM and a heat flux of 1.6 W/cm2. This system will extend the applicability of air cooling to power levels generally considered to lie in the domain of liquid cooling, and thus to the ATM switching nodes for B-ISDN.

A local area network (LAN) can now provide high-speed data communications in a local area environment to establish distributed processing among personal computers and workstations, and the need for interconnecting LANs, which are geographically distributed, is naturally arising. Asynchronous Transfer Mode (ATM) technology has been widely recognized as a promising way to provide the high-speed wide area networks (WAN) for Broadband Integrated Services Digital Network (B-ISDN), and the commercial service offerings are expected in the near future. The ATM network seems to have a capability as a backbone network for interconnecting LANs, and the LAN interconnection is expected to be the first service in ATM networks. However, there remain some technical challenges for this purpose; one of the main difficulties in LAN interconnection is the support of connectionless traffic by the ATM network, which is basically a connection-oriented network. Another one is the way of achieving the very high-speed data transmission over the ATM network. In this paper, we first discuss a LAN internetworking methodology based on the current technology. Then, the recent deployments of LAN interconnection methods through B-ISDN are reviewed.

Connectionless service for LANs interconnection will be provided in ATM networks at an early stage of B-ISDN era. This service will be provided on connection oriented mode at ATM technology. To perform this service, ATM connections using the dedicated bandwidth for this service are established semi-permanently between the nodes accommodating LANs. On these ATM connections, connectionless service among LANs is provided. It is important for private networks to utilize this bandwidth efficiently for reducing communication cost. In this paper, the architecture to provide connectionless service in private networks is described. Next, the allocation schemes of the bandwidth for this service and their performance are considered. We discuss the following schemes and compare them. One scheme is to establish semi-permanent ATM connections between the nodes with LAN interfaces. The bandwidth for each connection is individually assigned between these nodes. In another scheme, CLSFs (Connection-Less Service Functions) are introduced for connectionless service and connections are established via CLSFs. We show the latter scheme is superior because it brings out the effectiveness of statistical multiplexing of ATM technology and it leads to the reduction of the allocated bandwidth.

This paper deals with methods for interconnection between two local private networks that are geographically separated. A scheme is first presented to chain low bit-rate physical circuits into one logical circuit, over which ATM cells are transmitted as if there is one circuit with a high bit-rate capacity. In particular, use of existing low bit-rate circuits, e.g., 384/1536 kbit/s PDH leased line services and N-ISDN switched channels, is considered. The paper discusses two methods to permit chaining of physical circuits, and identifies their advantages and applications. By using the ATM-based circuit-chaining method, dynamic capacity control of the interconnection is then introduced with the use of an ATM-based rate adaptation. This is intended to provide a flexible and cost-effective capacity control compared to the existing TDM-based control. It is also possible to realize non-stop operation of changing capacity by establishment and release of chained circuits, which will lead to high reliability and robustness of private networks. Finally, delay characteristics introduced by the method are evaluated based on a computer simulation which gives a short and acceptable delay.

The demand for large-capacity photonic switching systems will increase as regular broadband ISDN (B-ISDN) spreads and full-motion video terminals replace telephones. Large-scale and economical optical fiber transmission lines have been built based on time-division (TD) multiplexing. To reduce costs, it is important to increase the channel multiplexity of both transmission and switching systems by using TD and wavelength-division (WD) or frequency-division (FD) technologies. We surveyed photonic switching systems' architecture and switching network structures. Switching can be divided into circuit or synchronous transfer mode (STM) switching, and asynchronous transfer mode (ATM) switching. A variety of photonic STM and ATM switching systems based on the two switching technologies have recently been proposed and demonstrated.

Broadband ISDN, using asynchronous transfer mode, are expected to carry traffic of different classes, each with its own set of traffic characteristics and performance requirements. To achieve the quality of service in ATM networks, a suitable buffer management scheme is needed. In this paper, we propose a buffer management scheme using a priority service discipline to improve the delay time of delay-sensitive class and the packet loss ratio of loss-sensitive class. The proposed priority scheme requires simple buffer management logic and minor processing overhead. We also analyze the delay time and the packet loss ratio for each class of service. The results indicate that the required buffer size of the proposed priority scheme is reduced and the delay time of each class of service is controlled by a parameter. If the control parameter is appropriately chosen, the quality of service of each class is improved.