Principles of TINA

TINA is intended to be applied to all parts of telecommunications and information systems: for example, terminals (personal computers, etc.), transport servers (switching systems, routers, etc.), service servers (VoD, web, etc.) and management servers (authentication, billing, etc.).
Reference Points are defined to specify conformance requirements for TINA products.
The architecture is based on four principles.

1) Object-oriented analysis and design,
2) distribution,
3) decoupling of software components,
4) separation of concern.

The purpose of these principles is to insure interoperability, portability and reusability of software components and independence from specific technologies, and to share the burden of creating and managing a complex system among different business stakeholders, such as consumers, service providers, and connectivity providers.

Object-oriented analysis captures the complexity of a system from different angles and breaks it down into a set of models. (Refer to the Model and Reference Point Section).
Distribution of service software components over different parts of the network accommodates traffic characteristics, network load or survivability, and specific customer demand. The distribution is supported by the Distributed Processing Environment (DPE).
Software components are decoupled from each other so that a change in one component due to a change in underlying technology (standards, languages, programs, materials, networks, etc.) would not affect other components.

TINA provides for two major separations of concern (Fig. 1).
The first separation is between applications and the environment (i.e., DPE) on which they run. The second is separation of applications into the service-specific part and the generic management and control part.
The latter interacts with transport and other elements.
According to the separation principles, TINA is divided into the following three
(sub-)architectures:

• Computing Architecture defines modeling concepts and the DPE. The DPE resides in heterogeneous pieces of equipment, and, by hiding their distribution, makes them function as a single system for applications. The TINA DPE is based on OMG's CORBA with adaptation for telecommunications requirements.
• Service Architecture defines a set of principles for providing services.
  It uses a notion of session to offer a coherent view of the various events and relationships taking place during the provision of services.
• Network Architecture describes a generic, technology-independent model for setting up connections and managing telecommunication networks.

TINA defines a set of reference points, derived from the TINA Business Model. Conformance to these reference points insures interoperability between TINA products. TINA captures the complexity of a system using a number of models, similar to ODP viewpoints.

• The Business Model describes the different parties involved in service provisioning and their relationship to each other. A small number of roles are defined, which reflect the major business separations of a complex telecommunications and information market: consumer, retailer, broker, third party service provider, content provider, and connectivity provider. The Reference Points comprises a set of interfaces describing the interactions taking place between these roles. For example, the Reference Point "Retailer" describes the relation between the consumer and the retailer.
• The Information Model describes information-bearing entities, their relations to each other, and the constraints and rules governing their behavior. These are described using the OMT modeling techniques.
• The Computational Model describes computational objects and their relations. The Object Definition Language (ODL) has been developed to help define computational objects. It is an extension of OMG's IDL.

Computing Architecture defines modeling concepts and the DPE.
The DPE is the environment that supports realization of applications described in the computational model (Fig. 2).
The DPE assures a certain number of distribution transparencies (e.g., location, access). Thus, computational objects for applications interact with each other via the DPE without the need to consider details of the native computing and communications environment (NCCE).
The DPE services and facilities can be distributed over multiple nodes connected by the Kernel Transport Network (kTN). Both DPE nodes and kTNs are logical entities and can be implemented in a variety of ways.
The TINA DPE is based on OMG technology such as CORBA. It extends OMG technology to provide for functions specific to telecommunications, such as stream ( i.e., continuous information flows) communications, the composition of multiple interface objects (e.g., one interface object for usage and another for management) and notification service. Some of these ideas are being introduced to OMG's Telecommunication Domain Task Force, which is in the process of adopting it. TINA-C aims at maximum alignment of its DPE specifications with those of OMG.

The Service Architecture defines a set of principles for providing services. It is based on the following principles:

• The notion of session is used. It is an important original concept of TINA. A session represents the information used by all processes involved in the provision of a service for a certain duration. For example, in a multipoint conference, the information about connections, charging and user profiles may change during the conference as participants join and leave. The session helps keep such information coherent throughout the conference. A session can also be very simple (e.g., a web search). The notion of session is further refined to allow for separation of access, service usage, and communications.
• Objects are separated into generic objects (common to all services) and service-specific objects (representing service logic, data, management, etc.),

Several sessions are defined, corresponding to different types of activities (Fig. 3):

• The access session corresponds to the establishment of the terms and conditions of the session (e.g., authentication, selection of service profile) during the connection of a user to a system. It allows the user to start service sessions, combine sessions, and participate in several services.
• The service session corresponds to the provision of the service itself (e.g., moderated multi-party conference with information retrieval capabilities) and insures overall coherence of control and management. It is divided into:
• the user service session that manages the state of each user's activity (local view) and resource attributes (e.g., charging context, current page),
• the provider service session that contains the service logic and offers the functions allowing the user to join a session, to be invited in a session, etc. (global view).
• The communication session provides an abstract view of the actual transport network connections. This is described in the Network Architecture below.

The Network Architecture defines a generic technology-independent model for setting up connections and managing telecommunications networks. It inherits concepts used in ITU-T and other standards bodies. It extends these concepts to integrate network control and management software for different network technologies.
The Network Architecture has the following three layers:

• The Communication Session layer provides service-independent interfaces for the service components to manage end-to-end communication at an abstract level. Therefore, this layer concerns both network and terminal parts of the communication.
• The Connectivity Session layer abstracts all the different network technologies and provides technology-independent interfaces for the communication session layer to interconnect network-level termination points. It also handles interworking between different network technologies.
• A Layer Network is an abstract generalization of each specific network technology. It deals with the setup and management of connection within a specific network technology.

For more information, look at Specifications.