Asynchronous transfer mode (ATM) networks are aimed at supporting a variety of services, including voice, video and data, while allowing the users to efficiently share network resources. Efficiency and flexibility in managing the resources are essential to meet the different quality requirements of these services. Network resources include trunk bandwidth, switching capacity, and buffer space in the ATM cross-connects. Sharing these resources is a complex task, since different applications have different traffic characteristic and quality of service (QOS) requirements. For example, video is characterized by continuous bit rate (CBR) or variable bit rate (VBR) traffic, while voice and data are bursty in nature. The quality requirements also depend on the type of service under considerations—some services, like video and voice, have strict end-to-end delay requirements, whereas others may be carried on a best-effort basis.

I n this paper, we propose a media access protocol for an optical star-coupler network in which the stations are equipped with fixed transmitters and tunable receivers and use wavelength-division multiplexing. The protocol, called the Group T D M scheme ( G T D M ) , is a reservation-based sch,eme in which the nodes are assembled into groups. The nodes within a group communicate with the nodes of other groups using the T D M scheme, bul they make reservations using the Dynamic Allocation Scheme (DAS), which has been proposed earlier. G T D M tries to overcome the drawbacks of T D M (unnecessary high delays under low load) and D A S (high signaling costs) by providing a trade-off between packet delay and signaling costs. Analytical and simulation results show that G T D Mcan achieve good performance with low signaling complexity. It is also shown to perform better than Hybrid T D M (HTDM), a hybrid of T D M and DAS, for a given signaling complexity.

No abstract

Each node of the Intel iPSCI2 and iPSCl860 is capable of concurrent bidirectional communication, that is, each node can concurrently send and receive a message. However, link contention created by the long message communication protocol usually interferes with this source of concurrency so that half of the total capacity of the communication network is wasted. This intelference can be overcome by synchronizing communicating pairs of nodes. The cost of algorith,ms for the simple exchange, the one-to-all broadcast, the allto-all broadcast, and the complete exchange operations can be significantly reduced, compared to the cost of their usual implementations, by exploiting this approach. Performance measurements of algorithms for these communication problems support these observations.

No abstract