Communication in all-optical networks requires novel routing paradigms. The high bandwidth of the optic fiber is utilized through wavelengthdivision multiplexing: a single physical optical link can carry several logical signals, provided that they are transmitted on different wavelengths. We study the problem of routing a set of requests (each of which is a pair of nodes to be connected by a path) on sparse networks using a limited number of wavelengths, ensuring that different paths using the same wavelength never use the same physical link. The constraints on the selection of paths and wavelengths depend on the type of photonic switches used in the network. We present eflicient routing techniques for the two types of photonic switches that dominate current research in all-optical networks. Our results es-

We survey the theoretical results obtained for wavelength routing in all–optical networks, present some new results and propose several open problems. In all–optical networks the vast bandwidth available is utilized through wavelength division multiplexing: a single physical optical link can carry several logical signals, provided that they are transmitted on different wavelengths. The information, once transmitted as light, reaches its destination without being converted to electronic form in between, thus reaching high data transmission rates. We consider both networks with arbitrary topologies and particular networks of practical interest.

Media access control protocols for an optically interconnected star-coupled system with pre-allocated Wavelength Division Multiple Access channels are introduced and compared. The photonic network is based on a passive star-coupled configuration where high topological connectivity is achieved with low complexity and excellent fault-tolerance. The channels are pre-allocated to the nodes with the proposed approach, where each node has a home channel it uses either for data packet transmission or data packet reception. This approach reduces the resulting system complexity since both tunable transmitters and tunable receivers are not required, and also has the advantage of being applicable to systems where there are many more nodes than wavelength channels. The performance of a generalized random access protocol is compared to an approach based on interleaved time multiplexing. Semi-markov analytic models are developed to investigate the performance of the two protocols. The analytic mode...

Wavelength conversion has been proposed for use in wavelength-division multiplexed networks to improve efficiency. This study highlights systems challenges and performance issues which need to be addressed in order to incorporate wavelength conversion effectively. A review/survey of the enabling technologies, design methods, and analytical models used in wavelengthconvertible networks is provided.

A passive-star-based, broadcast-and-select, local lightwave network which can support a limited number of WDM channels (on the order of ten), but serve a much larger number of nodes (a few tens or hundreds), is considered. Each node is equipped with one tunable transmitter and one fixed receiver, and each WDM channel is operated in a TDM fashion for carrying packet traffic. An important challenge is to allocate bandwidth to the node pairs when traffic flow between them is non-uniform, while also accommodating an important device constraint, viz. large transceiver tuning latency. Under such a scenario, this paper addresses the problem of determining optimum transmission schedules which can, in turn, minimize the network-wide mean packet delay. Specifically, a novel suite of algorithms for load balancing on a WDM/TDM local lightwave network are developed. Our approach exploits well-known results from scheduling theory to create good transmission schedules on such a network. The algorithm...

This paper examines wavelength translation in alloptical wavelength-routed networks. Previous studies [4-7] have shown that wavelength translation can improve the blocking performance of these networks. However, all previous work [4-7] has assumed that wavelength translators can translate from any input wavelength to any output wavelength. In contrast, alloptical wavelength translators demonstrated in the laboratory to date [9-11] are, in general, only capable of limited translation. In this paper we assess, for the first time, the network performance improvements offered by realistic all-optical wavelength translators with limited translation range. In particular, we consider all-optical wavelength translators based on four-wave mixing in semiconductor optical amplifiers. Using a simple model for their function, we consider the blocking performance of two-hop and multiple-hop paths, and unidirectional ring and mesh-torus networks. In all the cases we consider, significant improvement in the blocking performance of the network is obtained when limited-range wavelength translators with as little as one quarter of the full range are introduced. We also find that almost all of the network performance improvement offered by an ideal wavelength translator can be gained from a translator with only half of the full translation range. 1

A multiprocessor system with a large number of nodes can be built at low cost by combining the recent advances in high capacity channels available through optical fiber communication. A highly fault tolerant system is created with good performance characteristics at a reduction in system complexity. The system capitalizes on the optical self-routing characteristic of wavelength division multiple access to improve performance and reduce complexity. This paper examines typical optical multiple access channel implementations and shows that the star-coupled approach is superior due to optical power budget considerations. Starcoupled configurations which exhibit the optical self-routing characteristic are then studied. A hypercube based structure is introduced where optical multiple access channels span the dimensional axes. This severely reduces the required degree since only one I/O port is required per dimension, and performance is maintained through the high capacity characteristics of...

Scalable, hierarchical, all-optical WDM networks for processor interconnection in multiprocessor systems have been recently considered. The principal objective of this paper is to introduce an access protocol for this type of network which supports a distributed shared memory(DSM) environment. The objectives of the protocol are reduced averagelatency per packet, support of broadcast/multicast, collisionless communication, and exploitation of inherent DSM traffic characteristics. The protocol is based on a hybrid approach that combines reservation access and pre-allocated reception channels for a WDM system. The proposed approach trades maximum capacity for reduced communication latency to improve system response. The performance of the protocol is analyzed through semi-markov analytic and simulation models with varying system parameters such as number of nodes and channels. The performance of the new protocol is compared to a TDM-based protocol and their relative merits are examined. ...

This paper introduces two hierarchical optical structures for processor interconnection and compares their performance through analytic models and discrete-event simulation. Both architectures are based on wavelength division multiplexing (WDM) which enables multiple multi-access channels to be realized on a single optical fiber. The objective of the hierarchical architectures is to achieve scalability yet avoid the requirement of multiple wavelength tunable devices per node. Furthermore, both hierarchical architectures are single-hop: a packet remains in the optical form from source to destination and does not require cross dimensional intermediate routing. The first structure is physically hierarchical but wavelength flat: all nodes share the same wavelength space. The second structure is a wavelength multiplexed hierarchical structure with wavelength channel re-use at each level, allowing it to be scaled to very large system sizes. It employs acousto-optic tunable filters in conjunc...

All-optical WDM networks are fast becoming the natural choice for future backbone. In this paper, we establish the efficiency of multicasting over unicasting in all-optical WDM networks, assess the usefulness of wavelength conversion for multicasting, and explore the issues related to the splitting (or copying) capability of the nodes. The comparison between multicasting and unicasting is based on the number of wavelengths as well as the amount of bandwidth required for a given set of multicasting sessions. For each multicasting session, a source-specific multicasting forest (or trees) is constructed first, taking into account the sparse splitting capability of the nodes in the network. Then, each multicasting tree is partitioned into segments according to the sparse wavelength conversion capability of the nodes on the tree such that each segment needs to be assigned the same wavelength. Simulation results obtained for a practical network such as NSFNET and randomly generated networks ...