First-Generation Platforms: Proprietary, In-House Design and Development
In the 1980s and early 1990s, the pace of telecom service development was slower than it is today. Back then, Network Equipment Providers (NEPs) were vertically integrated organizations, developing their own service platforms and shopping around for components to build the requisite sub-systems. The components they selected were positioned as “best in class,” a claim based on cost, performance, feature set, availability, supplier stability, and service. The time for evaluation and selection of these components would typically range from one to six months.
“Best in class” is a slick marketing phrase, but it downplays — even disguises — the issues that arise when a single NEPs conducts all development and integration tasks in-house. These issues involve piecing together numerous third-party products. The process begins with the integration of line cards and protocol stacks, and is followed by the development or integration of a third-party database into high availability (HA) software, the integration of HA software with the hardware, and finally the integration of HA software with the protocol stacks. The time required to do all of this would typically last up to 18 months, and the building of the application could start only after the initial integration was complete.
Furthermore, the management phase — which involved administrating the hardware platform, stacks, line cards, HA software, database, and application layer — could also take 18 months or more. All told, a NEPs could spend 3-4 years building a workable proprietary telecom platform — a platform that would quickly become obsolete as the speed with which operators had to respond to customer demands for more flexible and dynamic services increased each year by what felt like orders of magnitude.
Second-Generation Platforms: Pre-Integrated, Semi-Standard Frameworks
As the pitfalls of first-generation platforms jeopardized the NEPs’ financial bottom lines, it became clear that time-to-market could be reduced if two or three suppliers collaborated in order to pre-integrate sub-systems. As a result, vendor selection criteria shifted to favor the ability to design and produce pre-assembled, pre-integrated, and differentiated functional elements.
This approach also allowed vendors to focus on discrete elements of the supply chain. For instance, as shown in the figure below, Supplier #1 might provide upper-layer network protocol stacks running on the intelligent carrier board while Supplier #2 might provide lower-layer network protocol stacks running on the I/O card. Supplier #1 could also provide management and HA software, as well as integration expertise for the sub-systems. Either vendor would offer convergence-layer software, which enables compatibility with third-party upper-layer protocol stacks and thus eliminates that integration step for the NEP. As the responsible party for the entire end product, the NEP would oversee the vendors’ work on the sub-systems while also developing the application-layer software that gives the system its unique functionality, such as Home Location Register (HLR), Mobile Switching Center (MSC), Softswitch, Signaling Gateway, etc.
The development of these second-generation platforms coincided with a dramatic rise in bandwidth and computing resources. The increasing power of central processing units and digital signal processors eliminated the need to hardwire functionality in order to obtain required performance. Instead, performance could be implemented in software, allowing the same boards to be used for different tasks — which enabled economies of scale to be realized and in turn reduced prices while increasing reliability. As a result, this advance in technology further paved the way for a move away from proprietary solutions toward standards-based platforms such as CompactPCI and AdvancedTCA.
Third-Generation Platforms: Standards-Based, Deployment-Ready Solutions
Despite the decrease in time-to-market afforded by second-generation platforms, many NEPs still found the pace of development too slow to meet the surging demands of their customers. The integration of multiple sub-systems from various vendors remained complex and time-consuming and, ultimately, too difficult to manage. This distraction prevented them from focusing solely on their application-based core competencies, thereby weakening their value proposition to the carriers.
In order to overcome these challenges, it became clear that true application-ready platforms were necessary to further increase the speed of network deployments and service enhancement. NEPs recognized the advantage of working with fewer sub-system suppliers in order to acquire what were quickly becoming standards-based platforms for hosting their wireless applications. The network element supply chain had thus evolved to the point where a small collection of horizontally-integrated suppliers producing standards-based products yielded the most rapid and robust service delivery platforms available. The total had become greater than the sum of the parts.
As a result, the concept of application-ready platforms emerged as the ideal solution to provide carrier-class, optimized performance while facilitating speed and overall ease of development. As shown in the figure below, with pre-integrated platforms NEPs gained the ability to source “out-of-the-box,” deployment-ready platforms to deliver whole product solutions swiftly and confidently — a necessity in today’s telecom market, which demands new value-added services at a rapid-fire pace. As a result of these application-ready platforms, the challenges of time to market and speed of delivery can be overcome, thereby further accelerating, converging and changing the communications industry for the better.
