 In order for the transition of IP networks to NGN to be fully successful, we have to understand, we have to treat this IP network like a career grade network. A career grade network is the one that involves certain guarantees. So, IP based environment has to adopt to the career grade open environment requirements. We will see what necessitates openness, what are the compliance requirements. We look at the ecosystem of this open environment. We look at different kinds of providers and finally we look at the career grade open environment reference model. So, why do we need openness? Traditional networks like PSTN, PLMN and other kinds of networks like POTS, they lack openness in their solution because they had been closed systems to begin with. But we can't afford that to happen in IP networks especially once we want to transition into next generation network. So, there is a concept that we need to understand which is what is going to be the relationship of the paraphernalia, the equipment, the hardware, the software, all known as the COTS commercially off the shelf components which are readily available in the market. How can we allow these components to be reutilized, re-engineered for the creation of different platforms, networks and applications all complying to NGN? For that, career grade open environment reference model is defined. It addresses all the concerns which are based on COTS component and corresponding standards which are accepted by the industry for implementing NGN. The compliance requirements that COTS have to fulfill are defined by the career grade requirements. The first one is, for instance, high performance. How many massive simultaneous sessions could be handled? What is the availability? Five nines? Five nines actually turns out to be like five minutes downtime a year or even six nines or maybe higher. Scalability, that is, to move from small network to a very large network with a lot of diversity in hardware, in coverage, in user base, etc. Then the hardware changes. How COTS have to be cognizant or aware of the hardware changes. Then software updates. Uniform and efficient management of COTS is going to be a very important concern under this CGOE reference model. Then protocol adaptation, that is, how existing protocols could be as smoothly as possible tailored to become workable with COTS. Then the controlled life cycle, that is, when a certain COTS is commissioned, is used, then it needs to be retired, that is, it is reaching the end of its life cycle. Then without having any impact on a live network performance, how can we remove it and plug in a new commercially off-the-shelf component? Then this requirement on part of COTS to be rapidly deployable and should be available for different rigorous testing conditions. And last but not the least, all of these requirements have to be met at extremely high cost efficiency. The COTS providers are not necessarily what we would assume as someone who sells equipment as in hardware or software like a vendor, but there is an abstraction to it. The abstraction actually is termed as the way we are looking at the providers in a more structured manner while keeping their internals as translucent as possible. So first we have the service providers. These are responsible for delivery of services to the end users. Then we have the solution providers, which deliver the solution building blocks, which essentially is used for providing these services. And the infrastructure that runs these services is managed by the technology providers to deliver the functional components to the service providers. The COTS ecosystem is a very important concept because with regards to career-grade open environment, we have to look at the coverage or scope of COTS. COTS essentially includes everything like networking platform, software and hardware platforms, then different functions for instance session management function, event handling functions. Then the overall management infrastructure and corresponding open interfaces. And very importantly, the security concerns and the career-grade performance ensuring functions. Now all these are embodied as the COTS ecosystem. We are now moving to the reference component model. The details of it can be found in text. But for brevity, we'll say that we have a layered architecture in which we have a top layer that is industry-specific career-grade open environment. That is whatever applications the industry is going to offer to the end users in NGN. Then the bottom is basically the industry agnostic environment. So industry in NGN doesn't bother about what hardware or what software or what network architecture is being deployed for as long as it is meeting the end user requirement specific features. So we have the application layer, we have the operating platforms and we have the server hardware. The applications could be categorized into further three types. We have the control plane applications like call processing language. We have the management plane applications for development and optimization of the components. And then we have the service plane applications. The service plane applications are the ones which are going to be incorporated like short message service center, content delivery network, web portals, file servers, etc. Then we have the operating platforms and server hardware. We'll just look at all this through this reference model. So we have the applications that already discussed here. For the sake of interest, we have the operations and management and provisioning of application services, basic networking application services and portal. Within the operating platform, we have a lot of cuts that could be incorporated at this particular layer. For instance, gateway protocol stacks, whenever a certain engine is going to talk to another engine, it's going to incorporate a gateway. So it means what protocol stacks are available for implementing gateway out there in the market. Then let's take the example of database middleware. How database middleware allows us to connect a certain component or group of components to the backend database. And then we have Java services, remote API services, data model services. Now this is a pretty exhaustive list, but for the sake of convenience, you can understand that operating platform is the environment in which a certain type of commercially off the shelf component is going to be executed. And then we have the hardware, the paraphernalia, including the servers, the networking equipment, router switches, their specifications, and their corresponding drivers which could be installed in a variety of end user equipment.