- 4.1 Conceptual Architecture Overview
- 4.2 EIA Reference ArchitectureArchitecture Overview Diagram
- 4.3 Architecture Principles for the EIA
- 4.4 Logical View of the EIA Reference Architecture
- 4.5 Conclusion
- 4.6 References
This chapter is from the book
This chapter is from the book
This chapter is from the book
4.4 Logical View of the EIA Reference Architecture
We are now ready to go from the Conceptual View of the EIA Reference Architecture to the first Logical View as shown in Figure 4.2. This diagram and the AOD described previously can be used to start the discussion about how one actually deploys these information components out into an enterprise.
)
Figure 4.2 Logical View of the EIA Reference Architecture
In Figure 4.2, we depict each of the major areas of the EIA Reference Architecture, including the general system and infrastructure components needed to operate and manage any IT landscape. In addition, a set of common requirements that straddle all the layers of the information architecture—Business Process Orchestration and Collaboration capabilities (not previously discussed), comprehensive Connectivity and Interoperability capabilities, and security-related requirements—all needed to manage and run any solution developed.
Each area is described in more detail in sections 4.4.1 to 4.4.7; this sets the scene for the following chapters on the Component and Operational Models.
The Component Model specifically describes how these functional aspects can be assembled to add value in any solution stack, and the Operational Model details how these functional components can be deployed onto physical assets to deliver the requirements (functional and non-functional) of the design. We now introduce each layer with a brief description.
4.4.1 IT Services & Compliance Management Services Layer
This layer covers the basic requirements upon which any solution must reside—it covers, for example, the set of tools and hardware required to configure, monitor, and manage the solution. Without this layer, no solution can be deployed because no infrastructure exists. Furthermore, there would be no way to manage the solution because there would not be features to monitor performance at the hardware, application, and business level. Also, change implementation and the ability to report back on any issues identified for problem resolution depend on this layer.
Recall that in the AOD we introduced a candidate ABB for Cloud Services and noted that we refine this positioning while detailing the EIA Reference Architecture. In the Logical View, we have the ability to take the first step into this refinement. Cloud Computing capabilities such as Elastic Capacity, a virtualized storage and IO layer, metering, billing, pricing, and monitoring are delivered through this layer. Other aspects of the Cloud Computing delivery model, such as multi-tenancy on the information layer, are discussed in the Component Model in Chapter 5.
4.4.2 Enterprise Information Integration Services
This Logical View shows how IT Architects can begin to manage the various data domains by first ensuring the data is managed, understood, and transformed in the EII layer. This layer is critical to the success of any enterprise-wide venture. It might start around a specific theme, for example, to solve a DW problem, to integrate several differing content repositories, or to manage data movement around the enterprise through an ESB. However, over time, this layer should be used to ensure all information integration workloads are managed through this layer to ensure consistency and re-use of assets.
As introduced in the capability section, EII covers the areas of discovery, profiling, cleansing, transformation, replication, federation, streaming, and deployment of information integration services in the IT environment.
4.4.3 Information Services
Information Services provide a uniform way of representing, accessing, maintaining, managing, and analyzing Structured Data and content across heterogeneous information sources. They are broken down into the following subcategories of logical services based on the data domain reference model previously introduced.
4.4.3.1 Metadata Services
Metadata Services are intended to unify the management of Metadata related to the information solutions supported. Logical and physical data models, data quality metrics, an enterprise business glossary, or Metadata supporting end-to-end data lineage and impact analysis are classified and managed with services in this domain. In the context of an SOA, message models associated with data flows and their service definition, along with service and process definitions, can be held and maintained through these services.
Various user roles perform through various tool operations on Metadata-related artifacts such as data models, data analysis assessment results, or data integration flow specifications. The tooling within this layer links Metadata to enable various different users to gain a 360-degree view of how data is being used and managed throughout the enterprise. For example, in a business glossary for a data model, technical terms can be linked to business terms. In a subsequent step, deriving a set of exposed services that define and enable consistent use of a business object such as customer is simplified. Metadata Services ultimately serve an enterprise by enabling trusted information by having the context available as Metadata and improving the consistency of data use wherever possible.
4.4.3.2 MDM Services
MDM Services manage the entities that are most critical to the success of an organization, such as customer, product, contract, or location. Even though the success of an organization relies on them, the Master Data itself is often inconsistent, inaccurate, incomplete, and distributed across application silos. With the deployment of MDM Services, the single version of truth for Master Data in a defined (logically) centralized repository and synchronizes the Master Data across the existing legacy environment. In this way, Master Data can be used by any business service. The service consumer can be confident that (security permitting) a valid set of data for that entity has been managed in a consistent way.
4.4.3.3 Data Services
Data Services provide access to Operational Data stored in any type of data storage. Data Services manage relational data—and if the underlying database supports it—XML data with XQuery support instead of just putting XML (similar to other Unstructured Data) into columns with a data type such as Binary Large Object (BLOB). Through Data Services, queries can be exposed.
Data Services can be divided into two categories: core data operations services including CRUD operations, and the associated housekeeping functions such as logging.
4.4.3.4 Content Services
Content Services manage Unstructured Data such as documents, media, or files. The scope of this layer includes basic CRUD operations and more advanced access mechanisms (content federation and search). The goal of the Content Services is to expose (single source or federated) content as a service and enable these services to be used within content centric workflows.
4.4.3.5 Analytical Services
Analytical Services are typically providing insight after Operational Data has been harmonized, cleansed, enriched, and aggregated in specialized systems such as DWs. For example, a data mining service can be embedded within a reporting service to deliver information to end users without them needing any skills in the use of the mining service employed. They enable the business to adapt to changing market dynamics and everyday operational disruptions. They support a holistic approach to business management enabling basic and deep analytics, aligned information objectives, role-based visibility, contextual insight, and in-time actions.
4.4.4 Presentation Services and Delivery Channels
The Presentation Services and Delivery Channel layer provides various channels for users to access the information and capabilities from the services delivered in the lower levels of the logical stack. This layer will be decomposed into server- and client-side components in the Component Model.
EIA demand ubiquitous access within the users’ familiar environment and across multiple delivery channels. Information must be available through servers, laptops, PDAs, or smart phones anywhere and anytime the user needs it. Therefore, it is supported by an enterprise platform, which delivers a consistent view of information to users at the right time, in the right format, and the right language. Users in this case are defined as humans, applications, and business processes.
4.4.5 Information Security and Information Privacy
Information Security and Information Privacy built into any solution are required for security and data assurance policies. They ensure an enterprise is compliant with all internal and (required) external policies. This in itself helps to maintain and even increase enterprise competitiveness.
Thus, these services are needed to create and maintain business-relevant, risk-appropriate solutions to cost-effectively meet and forestall security threats, evolve the enterprise to a posture of continual security and compliance readiness, and optimize response to regulatory, audit, and competitive pressures.
4.4.6 Connectivity and Interoperability
The Connectivity and Interoperability layer provides interoperability between the services within and beyond the enterprise. A typical instantiation of this layer is an ESB providing support for a variety of transport and communication protocols, interface mapping capabilities, and more interoperability functions.
4.4.7 Business Process Orchestration and Collaboration
The Business Process Orchestration and Collaboration layer provides two key capabilities. First, this layer provides end-to-end business process orchestration capabilities based on workflows containing, for example, automated tasks as well as human tasks. These features enable workflows to be driven across differing groups of users and systems to automate and streamline existing processes or to build or even outsource processes beyond the enterprise.
Second, this layer provides the capabilities so that all users in the enterprise can collaborate in an appropriate fashion. Typical examples of collaboration services include instant messaging and e-mail services.