PDF of this tutorial
Mediation is a data-translation methodology supported through proprietary point-to-point messaging services. Service providers use mediation to integrate real-time interactions between applications, network elements, and data sets—including customer care, provisioning and billing services. Many service providers, particularly CLECs, seek to expand the benefits of first-generation mediation services by adding new software technology to align mediation with IT business processes. To add this new capability, they are using a new category of server software.
Benefits and Limits of Traditional Mediation
Currently, service providers use a combination of third-generation languages (3GL) and various object technologies to build mediation services. Component-based frameworks such as common object request broker architecture (CORBA) or telecommunications management network (TMN) offer a way for service providers and equipment manufacturers to normalize network data into abstract formats. A service provider might use an object-oriented architecture for developing an interface to a billing system.
Open mediation models provide common translation between data sets and network elements, including application programming interfaces (APIs) supported through third-party applications and proprietary tools. This level of mediation has freed developers from the labor-intensive and time-consuming task of interfacing between service creation resources. Developers of third-party mediation solutions have used the benefits of open, object-based standards for telecommunications to enhance their ability to integrate "stovepipe" operations support systems (OSSs)/business operations systems (BSSs).
However, problems can arise when implementing a variety of technologies. For example, a CORBA–based provisioning system requires costly infrastructure integration to work with a fault-management system using Java. Additionally, existing solutions only address part of the service delivery problem. SCEs are designed to model only the network switching component of a service—not the billing and customer care systems earlier discussed. Traditional models for integration are static and require 3GL coding for change management. Therefore, unplanned interactions are impossible and require service downtime. The following changes to a network service might require additional custom software support for reintegration:
- operating system upgrades
- new database installation
- billing system upgrade
- addition of new software packages
- new service capabilities introduced
Because each piece of custom software must be supported as long it is in the network, each integration project brings with it a large, ongoing support cost. The limit of basic mediation is that it requires hand-coding interface changes each time there is a change, which is a frequent occurrence.
Using an Application Services Framework to Generate a Common Billing CDR
One of the most difficult areas for many service providers is the billing and customer care process. This is a result of the inherent complexity of handling customer-care functions that are shared across multiple applications while mediating large amounts of data—especially those that come from network CDRs. Convergent billing processes can be detained by implementation problems that limit the speed and cost advantage of deploying new services. Ongoing integration snarls necessitate manual processes, resulting in project delays and higher costs for ongoing support.
CLECs are among the most audible proponents of embracing an application services framework to add a common billing service model to their existing billing process. Typical CLECs must support and provision new services across multiple interconnection systems. Not only do they support smaller customer bases, they often must customize projects to suit the needs of a small groups of clients or even individual customers. Therefore, adaptable CDR formats are required for many services and interconnection functions.
As shown in Figure 4, mediation performs the consolidation of CDRs. This is a fundamental task required for offering basic multiservice network support. Convergent billing must be offered to consolidate the management of multiple service elements. In addition, mediation can be used to translate this CDR across different data models with applications while managing real-time interactions.
Figure 4. Billing Mediation to Create a Common CDR
How Application Services Framework Supercharges Mediation
Application services framework provides an automated approach to enhance mediation. When compared with traditional 3GL hard-coding approaches, application services framework offers significant savings of time and labor for developing new software. Application services framework speeds up changes to mediation and lowers cost. This is very attractive to service providers who are reducing the amount of infrastructure software they must otherwise develop and maintain in-house.
Simplification
In managing complex data models, application services framework creates a point-of-specification with a single external data model. This creates transparency between the heterogeneous set of APIs and execution runtime implementations such as CORBA, Java, SCEs, other proprietary EAI technologies, and Web servers. As a result, the tasks of integrating and updating interfaces are removed, making services easier to change and manage.
Particularly for CLECs on a tight budget that nevertheless seek to roll out new services and realize profits quickly, mediation using application services framework offers a valuable means of avoiding ongoing support complications. As managers of telecommunications integration projects assert, adding application services framework, rather than a 3GL method, leads to dramatic reductions in deployment times and lower cost for technology integration and continued support. This is by far the largest immediate savings that service providers cite, given that some project times are cut by more than half.
Performing in Web Time
In addition, system integrators have identified the unique memory-resident performance features that application services framework brings to their mediation projects. They indicate that real-time reliable cache performs well in maintaining data integrity, and these in-memory resident objects add the guaranteed application recovery capabilities of a true telecommunications service environment. The built-in scalability, high-availability, and fault-tolerance levels mean that the introduction of low-cost new services will not suffer from performance and scalability limits.
In managing billing systems, it is useful to make changes without taking services offline. For service providers, the ability to perform changes on the fly, with zero downtime, is another dramatic advance. For example, integrated with an existing billing system, application services framework makes it possible to perform modifications to existing fields or add new fields without interruption in the underlying billing application. According to service providers, this basic feature is important to adding novel and marketable services quickly.
In these situations, adding application services framework has gone beyond the mechanical task of consolidating the CDR; it has extended the utility and efficiency of a current mediation approach that was once cost-constrained to only the most readily updated interfaces with applications, network elements, and databases using object languages.
