Bellcore has built the SONET Toolkit decision support software to design robust fiber-optic networks that protect services against the consequences of a cable cut (link) or an equipment failure (node). The Synchronous Optical NETwork (SONET) makes these survivable designs cost-effective when planned carefully. The SONET Toolkit reads in data about the network, its embedded capacity, the available equipment, the customer demands, and the protection requirements for the services. It can also incorporate planning constraints. It produces an economic mix of SONET self-healing rings and other architectures that satisfy demand and protection requirements. Use of this software system by the Bellcore client companies has saved 10 to 30 percent in costs and orders of magnitude in time.

The introduction of SONET technology opens o p portunities for survivable network architectures, such as selfhealing rings, which can improve telecommunication reliability. This paper presents an algorithm for routing fiber around a ring in a network, when the network nodes, links, connectivity, and which offices are to be placed on that ring together are known. The algorithm aids automated survivable network planning. The algorithm was programmed in C, and run on a SPARCstation. Computation times on 47 examples of feasible and infeasible rings were reasonable. Overall, the average, minimum, and maximum runtimes were 0.41 sec., 0.06 sec., and 2.93 sec., respectively. Since the largest example network used in these results, 167 offices and 240 links, is the size of a typical large LATA network, the algorithm runs fast enough for the intended application. In most cases, the ring routing problem cannot be solved by traveling salesman algorithms. However, under certain conditions, the problem degenerates to the traveling salesman problem, and the ring routing algorithm degenerates to the nearest neighbor method of solving that problem.

A prototype software system that implements a methodology for the strategic planning of survivable interoffice networks is presented. The software system determines strategic locations and ring types for Synchronous Optical Network ring placement. Two types of survivable network architectures are considered-1 : 1 diverse protection and SONET self-healing rings. The software considers three types of SONET self-healing rings-unidirectional, 2-fiber bidirectional, and 4-fiber bidirectional. Hubbing is assumed in all architectures. Inputs include nodes, links, connectivity, facility hierarchy, and multiyear pointto-point demands, together with the costs of fiber material and splicing, route mileage (installation), and multiplexors and regenerators of different rates. The outputs are a set of near-optimal rings based on cost, specifying the ring types and rates, fiber span sizes and counts, regenerator locations and speeds, the topology (set of links to be used), and the network cost. In addition, the software outputs the time in the planning period that each ring and fiber span should be installed.

No abstract

REPORT DOCUMENTATION PAGE Form Approved OMB No. 074-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to This final technical report presents the results and conclusions from a set of research objectives posed by the Helios project. The objectives included: The study of all-optical LAN architectures and protocols; the establishment of a testbed for the purpose of demonstrating high-bandwidth applications and collecting and analyzing statistical profiles of their traffic; and the study of propagation of analog signals through all-optical networks. The report is structured as follows: Sections 2, 3 and 4 describe MCNC-RDI's work on the HiPeR-1 protocol and scheduler for broadcast WDM LAN architectures. Section 5 describes the application testbed established between MCNC-RDI, NCSU and UNC-CH using NCNI infrastructure for the purpose of accumulating statistical information about traffic demands of high-bandwidth applications. Section 6 describes the application cluster established within MCNC-RDI in order to test high-bandwidth/high-compute demand applications. Section 7 describes the research performed at Lucent Bell Labs concerned with transporting analog signals over DWDM optical links including physical layer impairments and adaptation layer studies. Section 8 describes the work performed jointly by UNC-CH and the University of Pennsylvania to establish a multimedia network testbed for telepresence applications. The Appendices present some of the technical details related to the HiPeR-1 scheduling protocol developed and implemented by MCNC-RDI.