November 1997

BY PETER CARLINO

Starting in the late 1980s, open programmable switches began to emerge as a byproduct of the increasingly competitive worldwide telecommunications marketplace. As interexchange carriers (IXCs), wireless carriers, and local loop providers around the world continue to fight their competitive battles, the wide-scale, low-cost deployment of enhanced service offerings becomes an all-important weapon. Open programmable switching technologies have proven popular because they provide scalability, economy, and a high degree of flexibility and programmability when compared with their traditional, closed counterparts. Today, open programmable switches are deployed by telecommunications providers in approximately 40 countries spanning six continents.

Open programmable switches are being used increasingly to build infrastructure from scratch in emerging networks such as wireless local loop, Personal Communication Services (PCS), and Specialty Mobile Radio (SMR). Traditionally, however, these switches have been used to augment existing infrastructure for the provisioning of new services. In the United States, for instance, the deployment of open programmable switches is enabling major providers who historically have competed on price alone, to establish differentiation and build subscriber loyalty through the deployment of an array of enhanced services. To date, the offerings deployed on the widest scale have included: Voice Activated Dialing, Cellular Call Screening, 800 Call Redirect, Credit and Debit Cards, “Follow Me” Services, and Information Retrieval.

The Problem: Today’s Model
Despite the success and the benefits of open programmable switches, solutions using this technology have yet to reach their full potential. The current state of today’s solutions built on programmable switches resembles the early days of personal computing. Specifically, the so-called “open” solutions in this space still require a high degree of configuration when integrated with products from multiple vendors that have myriad hardware and/or software interfaces. This solution process burdens application developers and inhibits carriers from implementing wide-scale deployments. For both of these groups, improvement will come in the form of universal standards, which will enable an entire new breed of “telco capable” CTI solutions to emerge.

For carriers, equipment pieced together from many vendors makes it impossible for them to predict with accuracy overall system performance or reliability. The lack of hardware and software standards means that multivendor solutions must be individually integrated, tested, and field-trialed to ensure that the given combination of components will meet the stringent reliability, management, and performance criteria that carriers typically require. In the present deployment of open programmable switches, carriers are mixing products from different vendors, some of which are certified for central office usage and some of which are not — a fact that is detrimental to overall reliability. In addition to performance issues, this integrated set of components by carriers can cause difficulty in the areas of network management and billing.

Software Developers
For software developers, the group responsible for driving the explosive growth of this market, the lack of standards is equally problematic. An application in an enhanced services environment may have responsibility for not only controlling a programmable switch, but for governing peripheral units such as speech recognition and voice response units as well. Because each of these hardware pieces is provided by different vendors, it is common that they will each have different Application Programming Interfaces (APIs). Specifically, each installation of an application may have to be tailored individually to operate in different hardware and network environments.

Just as in the PC analogy, widespread adoption of software is not possible until the underlying hardware can be addressed and controlled in a transparent manner. Because this is not the case today, developers often need to customize individual applications so that they will work at each site of deployment. The result is long, complicated development cycles and increased cost to the carrier. The creation of universal hardware and software interface standards will drastically reduce the required effort and lower costs.

The problems of these two groups are not isolated incidents, but rather, they are intertwined as a result of their symbiotic relationship. Specifically, carriers require innovative solutions which are predictable and reliable for mass deployments. Developers require mass deployments to remain profitable and innovative. This brings us to the need for standards.

Standards Are The Answer
The enhanced services industry cannot reach its true potential until a virtual ‘cookie-cutter’ approach is in place. This establishment will be made possible through the universal adoption and deployment of standards. To date, most of the progress in this area has been made on the customer premises side. De facto standards such as SCSA, MVIP, and TAPI have existed for several years. Recently, however, under the guidance of the Enterprise Computer Telephony Forum (ECTF) leading vendors have agreed to and adopted the standards H.100 and S.100, which represent significant progress in this area. The H.100 specification provides information to implement a CT bus interface at the physical layer for the PCI computer chassis card slot. The card-level definition of the overall CT bus specification will drive new applications and help open new markets by providing flexibility to equipment manufacturers, value-added resellers, system integrators, and those building computer based telecommunications applications.

S.100 defines a set of CT APIs that provides an effective way to develop CT applications in an open environment. It defines a client-server model in which applications use a collection of services to allocate, configure, and operate hardware resources. The S.100 integrates implementation details of call processing hardware and switch fabrics to enable portable applications to be written and furnishes these services via an operating system independent API that may be extended to support custom APIs. The S.100 also enables applications to be portable from one S.100 compliant platform to another.

S.100 uses a very efficient resource management scheme that isolates applications from resource implementation details. S.100 APIs take care of resource related and system administration details. In doing so, they allow application developers to stay focused on tasks unique to the application. To date, several manufacturers including Natural Microsystems and Dialogic have committed to the H.100 standard. Dialogic’s CT Media represents an example of an early implementation of an S.100-based software platform. While these activities represent significant progress, their roots are in the customer premise — not in the central office. H.100, for example does not address hot swapability of failed cards, a crucial requirement for service providers. S.100, while providing extensive media control, does not yet provide the discreet levels of call control necessary for use within service provider networks.

The challenge, then, is to take advantage of the momentum initiated by ECTF and evolve the standards to address the requirements of Central Office deployments. Once these standards are evolved and can become adopted, application developers and system integrators will be able to provide solutions to the service providers which not only address their product preferences, but also can be managed and administered more readily. As a result, service providers will be more comfortable in planning large scale deployments.

Another benefit provided by this integration is cost. Because carriers will eliminate redundant hardware components through the deployment of a tightly integrated solution, the overall footprint of the solution will be diminished, as will the price tag for such a system. Benefits will also be experienced in terms of support. Instead of having to call multiple vendors to diagnose potential problems, carriers will be able to call a single vendor to troubleshoot for the entire system. Since the S.100 API can control multiple devices, application writers now have the luxury of writing to a single API rather than through multiple ones. On the hardware side, service providers now have the freedom to add new hardware as they see fit, because the hardware adopts a standard platform. This enables them to scale their systems and add to them as needed without affecting the controlling software applications, which is unlike today’s situation.

What Does It All Mean?
When application developers are free to spend their time on innovation rather than customization and service providers confidently deploy solutions on a wide scale basis, the real beneficiaries are those of us in the user community. We will all be able to choose our service providers not only based on cost, but also on the services we receive from them which enhance our ability to communicate and ultimately raise productivity.

Peter Carlino, Summa Four director, Business Development and Corporate Communications, is responsible for developing strategic marketing plans for cellular and PCS markets. Summa Four, Inc., announced in June its intention to work with Dialogic Corporation on “Project Sigma” to develop a fully integrated, telco-capable, programmable switching platform based on industry standard hardware and software. As a fully-integrated, intelligent switching platform, this technology will enable telecommunications service providers to significantly reduce the cost and time-to-market of delivering advanced services.

To facilitate Project Sigma, Summa Four entered into a Joint Development Agreement with Junction, Inc. Junction is a Cupertino, CA-based development company whose principals have significant experience in the design and development of telephony solutions

Summa Four, Dialogic, and Junction will integrate CT Media, Dialogic’s implementation of the S.100 API, with ICS, Summa Four’s telco-capable call model. This integration is intended to produce a S.100 compliant API with a centralized resource management structure that is fault tolerant and highly distributed. They also plan to enable Dialogic’s DM3 family of media processing cards that conform to the ECTF’s H.100 bus standard to be fully integrated onto the open switch’s backplane. This integration is intended to eliminate overlapping hardware costs and reduce system footprint through tighter subsystem integration.