34.1 Overview

The TOGAF Architecture Development Method (ADM) provides a process lifecycle to create and manage architectures within an enterprise. At each phase within the ADM, a discussion of inputs, outputs, and steps describes a number of architectural work products or artifacts, such as process and application. The content metamodel provided here defines a formal structure for these terms to ensure consistency within the ADM and also to provide guidance for organizations that wish to implement their architecture within an architecture tool.

34.2 Content Metamodel Vision and Concepts

This section provides an overview of the objectives of the content metamodel, the concepts that support the metamodel, and an overview of the metamodel itself. Subsequent sections then go on to discuss each area of the metamodel in more detail. Contents of this section are as follows:

• Core content metamodel concepts (see 34.2.1 Core Content Metamodel Concepts) identifies the key concepts within the core content metamodel, including:
  • Core and extension content
  • Formal and informal modeling
  • Core metamodel entities
  • Catalog, matrix, and diagram concept
• Overview of the TOGAF content metamodel (see 34.2.2 Overview of the Content Metamodel) provides a high-level overview of the content of the metamodel.

34.2.1 Core Content Metamodel Concepts

A TOGAF architecture is based on defining a number of architectural building blocks within architecture catalogs, specifying the relationships between those building blocks in architecture matrices, and then presenting communication diagrams that show in a precise and concise way what the architecture is.

This section introduces the core concepts that make up the TOGAF content metamodel, through the following subsections:

• Core and Extension Content provides an introduction to the way in which TOGAF employs a basic core metamodel and then applies a number of extension modules to address specific architectural issues in more detail.
• Core Metamodel Entities introduces the core TOGAF metamodel entities, showing the purpose of each entity and the key relationships that support architectural traceability.
• Catalog, Matrix, and Diagram Concept describes the concept of catalogs, matrices, and diagrams.

Core and Extension Content

The role of TOGAF is to provide an open standard for architecture that is applicable in many scenarios and situations. In order to meet this vision, it is necessary to provide a fully featured enterprise architecture metamodel for content and also to provide the ability to avoid carrying out unnecessary activities by supporting tailoring.

The metamodel must provide a basic model with the minimum feature set and then support the inclusion of optional extensions during engagement tailoring.

The core TOGAF content metamodel and its extensions are illustrated in Figure 34-1.

The core metamodel provides a minimum set of architectural content to support traceability across artifacts. Additional metamodel concepts to support more specific or more in-depth modeling are contained within a group of extensions that logically cluster extension catalogs, matrices, and diagrams, allowing focus in areas of specific interest and focus.

All extension modules are optional and should be selected during the Preliminary Phase of the architecture development to meet the needs of the organization. Additionally, the extension groupings described by the content metamodel are only a suggestion and further tailoring may be carried out to suit the specific needs at the discretion of the architects.

This core and extension concept is intended as a move towards supporting formal method extension approaches within TOGAF, such as the method plug-in concept found within the Software Process Engineering Metamodel (SPEM) developed by the Object Management Group (OMG).¹

Core Metamodel Entities

The content metamodel uses the terminology discussed within the TOGAF ADM as the basis for a formal metamodel. The following core terms are used:

• Actor: A person, organization, or system that is outside the consideration of the architecture model, but interacts with it.
• Application Component: An encapsulation of application functionality that is aligned to implementation structuring.
• Business Service: Supports business capabilities through an explicitly defined interface and is explicitly governed by an organization.
• Data Entity: An encapsulation of data that is recognized by a business domain expert as a discrete concept. Data entities can be tied to applications, repositories, and services and may be structured according to implementation considerations.
• Function: Delivers business capabilities closely aligned to an organization, but not explicitly governed by the organization.
• Information System Service: The automated elements of a business service. An information system service may deliver or support part or all of one or more business services.
• Organization Unit: A self-contained unit of resources with goals, objectives, and measures. Organization units may include external parties and business partner organizations.
• Platform Service: A technical capability required to provide enabling infrastructure that supports the delivery of applications.
• Role: An actor assumes a role to perform a task.
• Technology Component: An encapsulation of technology infrastructure that represents a class of technology product or specific technology product.

A more in-depth definition of terms used within the content metamodel can be found in Part I, 3. Definitions.

Some of the key relationship concepts related to the core metamodel entities are described below:

• Process should normally be used to describe flow

  A process is a flow of interactions between functions and services and cannot be physically deployed. All processes should describe the flow of execution for a function and therefore the deployment of a process is through the function it supports; i.e., an application implements a function that has a process, not an application implements a process.

• Function describes units of business capability at all levels of granularity

  The term "function" is used to describe a unit of business capability at all levels of granularity, encapsulating terms such as value chain, process area, capability, business function, etc. Any bounded unit of business function should be described as a function.

• Business services support organizational objectives and are defined at a level of granularity consistent with the level of governance needed

  A business service operates as a boundary for one or more functions. The granularity of business services is dependent on the focus and emphasis of the business (as reflected by its drivers, goals, and objectives). A service in Service Oriented Architecture (SOA) terminology (i.e., a deployable unit of application functionality) is actually much closer to an application service, application component, or technology component, which may implement or support a business service.

• Business services are deployed onto application components

  Business services may be realized by business activity that does not relate to IT, or may be supported by IT. Business services that are supported by IT are deployed onto application components. Application components can be hierarchically decomposed and may support one or more business services. It is possible for a business service to be supported by multiple application components, but this is problematic from a governance standpoint and is symptomatic of business services that are too coarse-grained, or application components that are too fine-grained.

• Application components are deployed onto technology components

  An application component is implemented by a suite of technology components. For example, an application, such as "HR System" would typically be implemented on several technology components, including hardware, application server software, and application services.

Figure 34-2 illustrates the core entities and their relationships.

Catalog, Matrix, and Diagram Concept

The content metamodel is used as a technique to structure architectural information in an ordered way so that it can be processed to meet the stakeholder needs. The majority of architecture stakeholders do not actually need to know what the architecture metamodel is and are only concerned with specific issues, such as "what functionality does this application support?", "which processes will be impacted by this project?", etc. In order to meet the needs of these stakeholders, the TOGAF concepts of building blocks, catalogs, matrices, and diagrams are used.

Building blocks are entities of a particular type within the metamodel (for example, a business service called "Purchase Order"). Building blocks carry metadata according to the metamodel, which supports query and analysis. For example, business services have a metadata attribute for owner, which allows a stakeholder to query all business services owned by a particular organization. Building blocks may also include dependent or contained entities as appropriate to the context of the architecture (for example, a business service called "Purchase Order" may implicitly include a number of processes, data entities, application components, etc.).

Catalogs are lists of building blocks of a specific type, or of related types, that are used for governance or reference purposes (for example, an organization chart, showing locations and actors). As with building blocks, catalogs carry metadata according to the metamodel, which supports query and analysis.

Matrices are grids that show relationships between two or more model entities. Matrices are used to represent relationships that are list-based rather than graphical in their usage (for example, a CRUD matrix showing which applications Create, Read, Update, and Delete a particular type of data is difficult to represent visually).

Diagrams are renderings of architectural content in a graphical format to allow stakeholders to retrieve the required information. Diagrams can also be used as a technique for graphically populating architecture content or for checking the completeness of information that has been collected. TOGAF defines a set of architecture diagrams to be created (e.g., organization chart). Each of these diagrams may be created several times for an architecture with different style or content coverage to suit stakeholder concerns.

Building blocks, catalogs, matrices, and diagrams are all concepts that are well supported by leading enterprise architecture tools. In environments where tools are used to model the architecture, such tools typically support mechanisms to search, filter, and query the Architecture Repository.

On-demand querying of the Architecture Repository (such as the business service ownership example mentioned above) can be used to generate ad hoc catalogs, matrices, and diagrams of the architecture. As this type of query is by nature required to be flexible, it is therefore not restricted or defined within the content metamodel.

The interactions between metamodel, building blocks, diagrams, and stakeholders are shown in Figure 34-3.

34.2.2 Overview of the Content Metamodel

The content metamodel defines a set of entities that allow architectural concepts to be captured, stored, filtered, queried, and represented in a way that supports consistency, completeness, and traceability.

At the highest level, the content framework is divided up in line with the TOGAF ADM phases, as shown in Figure 34-4.

• Architecture Principles, Vision, and Requirements artifacts are intended to capture the surrounding context of formal architecture models, including general architecture principles, strategic context that forms input for architecture modeling, and requirements generated from the architecture. The architecture context is typically collected in the Preliminary and Architecture Vision phases.
• Business Architecture artifacts capture architectural models of business operation, looking specifically at factors that motivate the enterprise, how the enterprise is organizationally structured, and also what functional capabilities the enterprise has.
• Information Systems Architecture artifacts capture architecture models of IT systems, looking at applications and data in line with the TOGAF ADM phases.
• Technology Architecture artifacts capture procured technology assets that are used to implement and realize information system solutions.
• Architecture Realization artifacts capture change roadmaps showing transition between architecture states and binding statements that are used to steer and govern an implementation of the architecture.

A more detailed representation of the content metamodel is shown in Figure 34-5.

34.3 Content Metamodel in Detail

This section contains the following subsections:

• Core Content Metamodel (see 34.3.1 Core Content Metamodel) describes the metamodel entities that form the core content metamodel.
• Core Architecture Artifacts (see 34.3.2 Core Architecture Artifacts) lists the set of artifacts intended to accompany the core content metamodel.
• Full Content Metamodel (see 34.3.3 Full Content Metamodel) describes the metamodel entities that form extensions to the content metamodel.

34.3.1 Core Content Metamodel

Figure 34-6 shows the metamodel entities and relationships that are present within the core content metamodel.

34.3.2 Core Architecture Artifacts

35. Architectural Artifacts discusses in detail the way in which the underlying content metamodel can be used to present a set of catalogs, matrices, and diagrams to address stakeholder concerns.

The following set of artifacts are intended to accompany the core content metamodel:

34.3.3 Full Content Metamodel

When all extensions are applied to the core content metamodel, a number of new metamodel entities are introduced. Figure 34-7 shows which entities are contained in the core content metamodel and which new entities are introduced by which extension.

The relationships between entities in the full metamodel are shown in Figure 34-8.

34.4 Content Metamodel Extensions

As discussed earlier, the TOGAF content metamodel supports a number of extension modules that allow more in-depth consideration for particular architecture concerns. Figure 34-9 shows the core content metamodel and predefined extension modules.

During the Architecture Vision phase of a particular engagement, the scope of the engagement will be used to make a determination on appropriate extensions to be employed in order to adequately address the architecture requirements. For example, the scope of an engagement could be defined as core content, plus the governance extensions, as shown in Figure 34-10.

The following sections provide a more detailed description of the purpose and content of each of the extension modules.

34.4.1 Governance Extensions

Purpose

The governance extension is intended to allow additional structured data to be held against objectives and business services, supporting operational governance of the landscape.

The scope of this extension is as follows:

• The ability to apply measures to objectives and then link those measures to services
• The ability to apply contracts to service communication or service interactions with external users and systems
• The ability to define re-usable service qualities defining a service-level profile that can be used in contracts
• Creation of additional diagrams to show ownership and management of systems

This extension should be used in the following situations:

• When an organization is considering IT change that will result in a significant impact to existing operational governance models
• When an organization has granular requirements for service levels that differ from service to service
• When an organization is looking to transform its operational governance practice
• When an organization has very strong focus on business drivers, goals, and objectives and how these trace to service levels

The benefits of using this extension are as follows:

• Service levels are defined in a more structured way, with:
  • More detail
  • The ability to re-use service profiles across contracts
  • Stronger tracing to business objectives
• Impacts to operations and operational governance models are considered in a more structured way, with:
  • Additional diagrams of system and data ownership
  • Additional diagrams of system operation and dependencies on operations processes

In addition to the extensions described here, organizations wishing to focus on architecture governance should also consult:

• The COBIT framework for IT governance provided by the Information Systems Audit and Control Association (ISACA); refer to www.isaca.org
• The IT Portfolio Management Facility (ITPMF) from the OMG; refer to www.omg.org/spec/ITPMF

Required Changes to the Metamodel

Changes to the metamodel entities and relationships are shown in Figure 34-11.

Changes to the metamodel entities and relationships are as follows:

• Measure is added as a new entity that links objective and business service.
• Service Quality is added as a new entity that provides a generic service profile template to be applied to business services or contracts.
• Contract is added as a new entity that formalizes the functional and non-functional characteristics of a service interaction with other services, external applications, or users.

Changes to the metamodel attributes are as follows:

• Attributes are added for the new metamodel entities of Measure, Service Quality, and Service Contract

Additional diagrams to be created are as follows:

• Enterprise Manageability diagram

34.4.2 Services Extensions

Purpose

The services extension is intended to allow more sophisticated modeling of the service portfolio by creating a concept of IS services in addition to the core concept of business services. IS services are directly supported by applications and creating the layer of abstraction relaxes the constraints on business services while simultaneously allowing technical stakeholders to put more formality into an IS service catalog.

The scope of this extension is as follows:

• Creation of IS services as an extension of business service

This extension should be used in the following situations:

• When the business has a preset definition of its services that does not align well to technical and architectural needs
• When business and IT use different language to describe similar capabilities
• Where IT service is misaligned with business need, particularly around the areas of quality of service, visibility of performance, and management granularity
• Where IT is taking initial steps to engage business in discussions about IT architecture

The benefits of using this extension are as follows:

• Business services can be defined outside of the constraints that exist in the core metamodel. This allows for a more natural engagement with business stakeholders.
• IS services can be defined according to a model that maps closely to implementation, providing a more realistic solution abstraction to support IT decision-making.
• Business and IS service relationships show where the business view aligns with the IS view and where there are misalignments.

In addition to the extensions described here, organizations wishing to focus on services-centric architectures should also consult:

• The Service Component Architecture (SCA) specification developed by the Open Service Oriented Architecture (OSOA) collaboration; refer to www.oasis-opencsa.org/sca
• The Service Data Objects (SDO) specification developed by the Open Service Oriented Architecture (OSOA) collaboration; refer to www.oasis-opencsa.org/sdo

Required Changes to the Metamodel

Changes to the metamodel entities and relationships are shown in Figure 34-12.

Changes to the metamodel entities and relationships are as follows:

• IS Service is added as a new metamodel entity, extending business service.
• IS Service inherits all the relationships of a business service.
• A new relationship is created linking an IS service to a business service.

Changes to the metamodel attributes are as follows:

• IS Service is added as a new type of business service.

Additional diagrams to be created are as follows:

• Business Use-Case Diagram
• Organization Decomposition Diagram

34.4.3 Process Modeling Extensions

Purpose

The process modeling extension is intended to allow detailed modeling of process flows by adding events, products, and controls to the metamodel. Typically, enterprise architecture does not drill into process flow, but in certain process-centric or event-centric organizations it may be necessary to elaborate process in a much more formal manner using this extension module.

The scope of this extension is as follows:

• Creation of events as triggers for processes
• Creation of controls that business logic and governance gates for process execution
• Creation of products to represent the output of a process
• Creation of event diagrams to track triggers and state changes across the organization

This extension should be used in the following situations:

• Where the architecture must pay specific attention to state and events
• Where the architecture is required to explicitly identify and store process control steps; for example, to support regulatory compliance
• Where the architecture features critical or elaborate process flows

The benefits of using this extension are as follows:

• This extension allows detailed process modeling and the cataloging of process artifacts.
• May be used to support regulatory compliance activities.
• May be used to re-purpose legacy or non-architectural process decomposition analysis.

In addition to the extensions described here, organizations wishing to focus on process-centric architectures should also consult:

• The Business Process Modeling Notation (BPMN) specification, provided by the OMG; refer to www.bpmn.org
• The Software Process Engineering Metamodel (SPEM) specification, provided by the OMG; refer to www.omg.org/spec/SPEM

Required Changes to the Metamodel

Changes to the metamodel entities and relationships are shown in Figure 34-13.

Changes to the metamodel entities and relationships are as follows:

• Event is added as a metamodel entity, sitting between Actor, Process, and Service
• Control is added as a metamodel entity, relating to a Process.
• Product is added as a metamodel entity, linking Organization and Processes.

Changes to the metamodel attributes are as follows:

• Attributes are added for the new metamodel entities of Event, Control, and Product.

Additional diagrams to be created are as follows:

• Process Flow diagrams, showing the way in which business functions, events, controls, and products are linked to support a particular business scenario
• Event diagrams, showing events, were they are received from, and what processes they trigger

34.4.4 Data Extensions

Purpose

The data extension is intended to allow more sophisticated modeling and the encapsulation of data. The core model provides a data entity concept which supports the creation of data models, which is then extended by this extension to include the concept of a data component. Data components form a logical or physical encapsulation of abstract data entities into units that can be governed and deployed into applications.

The scope of this extension is as follows:

• Creation of logical data components that group data entities into encapsulated modules for governance, security, and deployment purposes
• Creation of physical data components that implement logical data components and are analogous to databases, registries, repositories, schemas, and other techniques of segmenting data
• Creation of data lifecycle, data security, and data migration diagrams of the architecture to show data concerns in more detail

This extension should be used in the following situations:

• Where the architecture features significant complexity and risk around the location, encapsulation, and management of or access to data

The benefits of using this extension are as follows:

• The structure of data is modeled independently from its location, allowing data models to be developed that span multiple systems without being tied to physical concerns.
• Logical groupings of data can be used to set governance, security, or deployment boundaries around data, providing a much more holistic appreciation of data issues surrounding the architecture.

Required Changes to the Metamodel

Changes to the metamodel entities and relationships are shown in Figure 34-14.

Changes to the metamodel entities and relationships are as follows:

• Logical Data Component is added as a new metamodel entity, encapsulating data entities.
• Physical Data Component is added as a new metamodel entity, extending Logical Data Component.
• A relationship is created between Physical Data Component and Application Component. If the infrastructure consolidation extension is applied, this should be to Physical Application Component.
• If the infrastructure consolidation extension is applied, Physical Data Components will have a relationship with Location.

Changes to the metamodel attributes are as follows:

• Attributes are added for the new metamodel entities of Logical Data Component and Physical Data Component.

Additional diagrams to be created are as follows:

• Data Security diagram
• Data Migration diagram
• Data Lifecycle diagram

34.4.5 Infrastructure Consolidation Extensions

Purpose

The infrastructure consolidation extension is intended to be used in landscapes where the application and technology portfolios have become fragmented and the architecture seeks to consolidate the business as usual capability into a smaller number of locations, applications, or technology components.

The scope of this extension is as follows:

• Creation of a location entity to hold the location of IT assets and external consumers of service
• Creation of logical and physical application components to abstract the capability of an application away from the actual applications in existence
• Creation of logical and physical application components to abstract product type from the actual technology products in existence
• Creation of additional diagrams focusing on the location of assets, compliance with standards, structure of applications, application migration, and infrastructure configuration

This extension should be used in the following situations:

• Where many technology products are in place with duplicate or overlapping capability
• Where many applications are in place with duplicate or overlapping functionality
• Where applications are geographically dispersed and the decision logic for determining the location of an application is not well understood
• When applications are going to be migrated into a consolidated platform
• When application features are going to be migrated into a consolidated application

The benefits of using this extension are as follows:

• Allows visibility and analysis of redundant duplication of capability in the application and technology domains
• Supports analysis of standards compliance
• Supports analysis of migration impact of application or technology consolidation
• Supports detailed architectural definition of application structure

In addition to the extensions described here, organizations wishing to focus on infrastructure consolidation should also consult:

• The Unified Modeling Language (UML), provided by the OMG; refer to www.uml.org
• The Systems Modeling Language (SysML) - www.sysml.org - which reduces the complexity and software engineering focus of UML for the purposes of systems modeling
• The IT Portfolio Management Facility (ITPMF) from the OMG; refer to www.omg.org/spec/ITPMF

Required Changes to the Metamodel

Changes to the metamodel entities and relationships are shown in Figure 34-15.

Changes to the metamodel entities and relationships are as follows:

• Location attributes on Organization, Actor, Application Component, Data Component, and Technology Component are enhanced to create a location entity within the metamodel.
• Application Components are extended to include Logical Application Components (a class of application) and Physical Application Components (an actual application).
• Technology Components are extended to include Logical Technology Components (a class of technology product) and Physical Technology Components (an actual technology product).

Changes to the metamodel attributes are as follows:

• Creation of attributes for the new Metamodel entities of Logical Application Component, Physical Application Component, Logical Technology Component, Physical Technology Component, and Location
• Removal of Location as an attribute of entities that have a location and replacement with a relationship with the Location entity

Additional diagrams to be created are as follows:

• Process/Application Realization diagram
• Software Engineering diagram
• Application Migration diagram
• Software Distribution diagram
• Processing diagram
• Networked Computing/Hardware diagram
• Communications Engineering diagram

34.4.6 Motivation Extensions

Purpose

The motivation extension is intended to allow additional structured modeling of the drivers, goals, and objectives that influence an organization to provide business services to its customers. This in turn allows more effective definition of service contracts and better measurement of business performance.

The scope of this extension is as follows:

• Creation of a new metamodel entity for Driver that shows factors generally motivating or constraining an organization
• Creation of a new metamodel entity for Goal that shows the strategic purpose and mission of an organization
• Creation of a new metamodel entity for Objective that shows near to mid-term achievements that an organization would like to attain
• Creation of a Goal/Objective/Service diagram showing the traceability from drivers, goals, and objectives through to services

This extension should be used in the following situations:

• When the architecture needs to understand the motivation of organizations in more detail than the standard business or engagement principles and objectives that are informally modeled within the core content metamodel
• When organizations have conflicting drivers and objectives and that conflict needs to be understood and addressed in a structured form
• When service levels are unknown or unclear

The benefits of using this extension are as follows:

• Highlights misalignment of priorities across the enterprise and how these intersect with shared services (e.g., some organizations may be attempting to reduce costs, while others are attempting to increase capability)
• Shows competing demands for business services in a more structured fashion, allowing compromise service levels to be defined

In addition to the extensions described here, organizations wishing to focus on architecture modeling of business motivation should also consult:

• The Business Motivation Model (BMM) specification, provided by the OMG; refer to www.omg.org/technology/documents/bms_spec_catalog.htm

Required Changes to the Metamodel

Changes to the metamodel entities and relationships are shown in Figure 34-16.

Changes to the metamodel entities and relationships are as follows:

• Driver, Goal, and Objective are added as new entities that link Organization Unit to Business Service.

Changes to the metamodel attributes are as follows:

• Attributes are added for the new metamodel entities of Driver, Goal, and Objective.

Additional diagrams to be created are as follows:

• Goal/Objective/Service diagram

34.5 Content Metamodel Entities

The following table lists and describes the entities within the content metamodel.

34.6 Content Metamodel Attributes

The following table shows typical attributes for each of the metamodel entities described previously.

34.7 Metamodel Relationships

Footnotes

1. Refer to www.omg.org/spec/SPEM.

return to top of page