Analysis and Design of Fully Shared Networks

Abstract CONTRIBUTION STATEMENT: Presents a new dynamic network concept which combines dynamic traffic routing performed by switch network elements with dynamic transport routing performed by digital cross-connect network elements. Describes mathematical models to design such fully shared networks (FSN), which achieve efficient and highly robust capacity design, and simplified network operation. Demonstrates significant network efficiency and performance improvements of FSN designs for normal daily traffic loads, network failures, unexpected traffic overloads, and peak day traffic overloads. Dynamic routing in telecommunications networks has been the subject of worldwide study and interest. Service providers, equipment providers, and academic institutions throughout the world have active research programs in this area. Dynamic routing networks have been in operation for nearly 10 years and many such networks are in the planning or deployment stage. First implemented during the 1980s, dynamic routing is now deployed in three major networks (AT&T USA, TCTS Canada, and NTT Japan) and has provided considerable benefits in improved performance quality and reduced costs [1]. These benefits have motivated the extension of dynamic routing to integrated networks with multiple classes-of-service and to networks with rearrangeable transport capacity, which is the subject of this paper. The fully shared network (FSN) is a new dynamic network concept which combines dynamic traffic routing with dynamic T1/T3 transport routing. FSN uses automatic control of DCS3/1 and DCS3/3 (digital cross-connect systems 3/1 and 3/3) to achieve dynamic bandwidth allocation of T1/T3 transport and switch capacity. It provides robust network design, T3-level capacity engineering, and automatic T1/T3 provisioning to achieve increased revenues and significant savings in capital and operations costs. The FSN concept builds on class-of-service dynamic routing capabilities in the switched network by allowing automatic implementation of self-healing network strategies such as T1 transport diversity, multiple homing, and T1/T3 restoration. Automatic control of DCS3/1 and DCS3/3 allows rapid provisioning and rearrangement of interswitch T1 capacity, access T1 capacity, and switching capacity in much shorter time periods than is possible today. An FSN design module receives daily network traffic data, designs and allocates T1/T3 transport capacity based on traffic levels, calculates an efficient rearrangement strategy, interfaces with the switches and DCSs through a control channel, and automates the provisioning and control of T1/T3 transport and switching capacity within the FSN.