Pushing the SONET Envelope with Auto-Provisioning

For most of its life, operational issues have dictated the evolution of the public switched telephone network (PSTN). For example, fiber-based synchronous optical network (SONET) technology was originally introduced to relieve the bandwidth shortage that resulted when asynchronous copper-based trunks reached their capacity. Under this network-focused model, there was little pressure to innovate and create new services. At the same time, network management was not end-to-end. Regardless, new network technologies, particularly SONET, prevailed.

Service providers have mostly abandoned the standard, pre-divestiture AT&T environment for network management and simple service delivery. Transmission and switching technologies have finally reached a point where they support advanced services as well as traditional services. Now, not only do optical networking technologies resolve bandwidth issues, but SONET in North America-and the comparable synchronous digital hierarchy (SDH) in Europe and other parts of the world-support high-speed requirements for video services. Contemporary network elements arrive with built-in intelligence that simplifies service provisioning. Vendor-supplied managers or customized, legacy OSSs generally control this intelligence to achieve remote provisioning via software.

However, the introduction of multi-vendor equipment and systems has accentuated many management problems for service providers. These problems have driven the evolution of the telecommunications management network (TMN) architecture and the automation and integration of operational business processes.

Figure 1 displays the TMN architecture, a standard comprising both the legacy and vendor-specific solutions required for network management. Information models based on standards (e.g., SONET’s GR 253) nevertheless often include vendor-specific extensions that step outside the standards, as well as homegrown applications, closed management systems and “black boxes” for mediation. This mix creates a complex network for managing and tracking logical and physical resources.

Service providers and vendors worldwide are building or buying solutions to fit into legacy and new ILEC and CLEC infrastructures (see Figure 2). They follow standards such as TMN and those developed by Bellcore to lower life-cycle development and maintenance costs. Such companies are implementing interfaces using data standards and standard management information bases (MIBs), along with local and long haul transport technologies. SONET, asynchronous transfer mode (ATM), digital loop carrier (DLC) and other standard information models help make the logical connections to physical resources.

Bell operating companies and competitive local exchange carriers have long used SONET as the de facto industry standard for transport networks. SONET networks utilize standard protocol interfaces to optimize bandwidth capabilities. These accommodate traffic from DS1 signals (1.544 Mbps) to higher speed optical carrier signals. In conjunction with fiber-optic media, the capacities of SONET transport networks are virtually limitless.

Many service providers are finding that provisioning services to SONET transport networks in a timely and efficient manner is a challenge, for several reasons:

1. Service providers tend to be trapped in a world of traditional provisioning trends. A philosophical paradigm shift must occur within an organization to transition from a narrow-band mindset to a more robust broadband point of view
2. SONET networks must be understood and modeled such that bandwidth services can be sold.
3. It is necessary to understand the roles of the provisioning organizations (e.g., a network operations center, or NOC) and their communication with field operations.
4. Solutions must allow the provisioning process as much automation as possible.
Understanding SONET Networks

Provisioning to SONET networks requires one to understand and accurately model the protocol interfaces. From a provisioning standpoint, SONET networks are nothing more than a higher-order multiplex network. Input signals range from DS1 to other OCn carrier signals (see Figure 3).

OCn carrier circuits traverse the network, providing high-speed bandwidth circuits to customers. A provider also uses bandwidth to extend network monitoring, expand interoffice facilities, etc. (see Figure 4).

A company may enhance its service offering by allowing traffic to be added or dropped at any point in the network without demultiplexing an entire signal (see Figure 5). This extends the service provider’s reach to its customers by using the topology of a wide-area network.

Viewed this way, provisioning to SONET networks becomes an exercise in managing capacity and accurately modeling equipment configuration. For field operations, this perspective reduces the possibility of work order errors during service activation.

Bridging the Gap

The real possibility of error lurks in the gap between service provisioning organizations and their field operations. Using existing operations support systems (OSSs), building new OSSs or buying commercially available ones is the weapon for bridging the gap. Regardless of whether an organization performs network provisioning centrally or locally, it is critical to the company’s success to employ its OSS infrastructure effectively to communicate to the field.

Traditional, legacy OSSs have had trouble handling the configurable nature of SONET equipment. Because of its complexity, field operations personnel would spend valuable time correcting the provisioning information or rejecting orders altogether. Either scenario increases activation intervals, which translate into lost revenue. Newer OSS solutions must be able to accommodate the range of interfaces that SONET can support (DS1, DS3, etc.) and communicate configuration information accurately and efficiently. In order to create the optimal solution for SONET provisioning, it’s important to understand that SONET network capacity (in the form of assignable facilities) is directly proportional to the equipment configuration (see Figure 6)

Provisioning to this network must break down the SONET facility to match its equipment configuration. This is done by managing capacity on the high speed (STS) level of the OCn carrier. The OCn carrier is essentially “sliced” into ST pieces. The pieces are distributed and channelized to accommodate the service orders that require them (see Figure 7). An OSS’s ability to deliver and communicate this information is critical to its success.

This modular form of facility creation allows an organization to more easily migrate into SONET topologies such as rings, single and dual-homed subtended rings, and dual-ring interworking. In addition, if critical information about configuration and connection assemblies appears on a work order, field forces spend less time researching orders and more time turning up service.

Flow-Through Provisioning-Service Deployment Nirvana

The last and most critical element for automating SONET provisioning is the ability to communicate this configuration and connection assembly information remotely. This is known as flow-through provisioning.

The latest telecommunications equipment supports a network operations center that handles monitoring and performance in a central location. This focal point can communicate with intelligent network elements via dial-up or telnet sessions. SONET provisioning uses this capability by allowing the organization to turn up service by communicating cross-connect information remotely. This approach eliminates the need for large field forces that traditionally performed these functions manually. In the example in Figure 8, specific types of cross-connects are identified for a digital cross-connect system (DCS) as it relates to a SONET ring.

If an OSS can present this information remotely in the correct format to a DCS or SONET node, service could be turned up with minimal manual intervention. Further, with genuine flow-through capability, the system requires only one touch from the provisioning group to design the circuit layout. Then the OSS and network management systems automatically perform remaining tasks and synchronize the data. This system ultimately shortens service activation intervals and reduces errors.

Requirements for OSSs are changing to support this enhanced process. These requirements include:
1. A graphical network design tool to aide in the search for a usable network route, especially for provisioners who are unfamiliar with the network

2. Auto-assignment of customers to bandwidth and channels to support true, automatic flow-through, so the provisioner does not have to choose the interface card and channel manually

3. A dynamic view of design information, to reduce errors commonly caused with textual descriptions

4. An integrated information flow-for example, a graphical representation of the design layout report/circuit layout report (DLR/CLR) information integrated with the administrative circuit detail, circuit design, and end user termination sections

5. A flexible configuration-for example, the ability to designate variable rates to channels

6. An ability to trigger events in the network management system to enable the provisioning process to support flow-through to the activation device.

The Moral of the Story

Automating their SONET provisioning is fast becoming a critical success factor for many service providers, whether they are providing the network backbone or simply relying on it for timely delivery of advanced products and services. One large service provider, based in Denver, offers network backbone as part of its core business strategy. By automating its SONET provisioning, it was able to measure its success by how many orders it could turn up against the SONET network in a day, versus how many days it took to provision a single order using its previous manual processes.

Originally, this service provider modeled its SONET network for linear configuration. Eventually it remodeled to support a ring configured, ADM-based network and as a result had to change its hardware configuration in the field. At the same time, in the back office, the service provider moved from multiple systems to a single system that supports point-and-click design functionality.

Organizationally, this service provider created efficiencies by cleaning up the hand-offs between responsible organizations such as the NOC and field operations. It also improved interdepartmental service level agreements by implementing stricter rules that mandated reduced handoff times. Some departments were also combined to eliminate redundant efforts and reduce the actual number of interdepartmental handoffs.

Overall, these changes resulted in a three- to four-fold, or 300 to 400 percent, improvement in service order provisioning efficiency; bringing its provisioning capability from one order per person, per day to 3 or 4 per person, per day. The service provider is also predicting a six-to-one (600 percent) improvement over its original processes based on efforts to improve design against access facilities management and by incorporating access design into the overall SONET configuration process. Special thanks to Arnie Cruz, Sr. Consultant, MetaSolv Software for his contribution to, and validation of, the concepts used in this article.