Architecture segregation

In 1992, Steven H. Spewak defined Enterprise Architecture Planning (EAP) as the process of defining architectures for the use of information in support of the business and the plan for implementing those architectures.

In highly distributed computing, a layered architecture is recommended for a simple reason--to allocate the different responsibilities of the system to the respective layers. With the same principle, Enterprise Architecture is built on the same layer design concept. It is inspired with the idea to execute the relevant processes and services of the layer and its related components. Take a look at the following image:

As defined in the preceding image, each layer, namely Business, Data, Application, and Technology, is designed to delegate its execution to the underlying layer. It means that the top layer, Business, is coarse-grained level, whereas the bottom layer, Technology, is a fine-grained level.

Business Architecture

Business Architecture is nothing but a blueprint of the enterprise. It helps you understand the organization and supports you to align the long-term strategic objectives and short-term tactical demands. Basically, it is the bridge between enterprise strategy and business functionality, as depicted in the following image:

As shown in the preceding image, Business Architecture has three dimensions at its core--Strategy, Operation, and Technology. The success factors of a Business Architecture are directly proportional to the business transformation, using strategy, a stable platform via technology, and exhibited excellence in the business operation.

In essence, the key aspect of the represented business by Business Architecture is tabulated as follows:

S.No.	Query raised	Target delivery	Sample
1	Who?	Stakeholders	End customers, senior managers.
2	Why?	Strategy Tactics	Vision, mission, objectives of the business.
3	How?	Initiatives Projects	Innovative development, operational excellence.
4	When?	Events	Business critical moments.
5	What?	Products Services	Manufacturing output, customer facing service.
6	Where?	Policies Regulations	Governing body, corporate policy, company process.
7	How well?	Metric Measurement	Financial report, revenue trend, profit sharing.

Business Architecture is directly based on business strategy. By design, it is the foundation for subsequent architectures, where it is detailed into various aspects and disciplines of the business.

One of the relevant Business Architecture case studies by Oracle is available at http://www.oracle.com/technetwork/articles/entarch/oea-case-study-brinks-2012-1883032.pdf.

In my view, an ideal business architect delivers the framework of services and artifacts, which enables customers to rapidly deliver quantifiable business value with realistic, technology enabled, business solutions.

Data Architecture

Data Architecture is the key contract between the business and technology in an enterprise. I experienced the value of Data Architecture during my tenure of the Enterprise Data Hub development initiative.

Data Architecture is designed in such a way that the real business data is handled smoothly across the layers of the enterprise architecture. It plays the key role/artifact to develop and implement governance supporting the enterprise data strategy. It collaborates/connects with the various enterprise objects, such as hardware, applications, networks, tools, technology choices, and data.

To support a variety of the commonly used enterprise applications and business improvement activities, the framework/layers of the Data Architecture is designed as follows:

As depicted in the preceding image, Data Architecture has three layers of components based on its operational strategy, namely Strategic, Tactical, and Operational. As self-described, most of the ground-level operations are executed in the lower components--Enterprise Application Integration (EAI), Business Intelligence (BI), and System Rationalization. Data is tactically architecture at the middle layer using the BPI program. The top layer of Data Architecture is getting involved in the Data Strategy of the underlying enterprise.

Let me illustrate with a real life example to easily understand enterprise data architecture. Our business use case is to build the inventory management system of a production factory. Consider the following image:

As depicted in the preceding image, the inventory management workflow is aimed towards the process of supervising and controlling stock items for the production in an efficient way. Let's get into the details of Data Architecture with this example.

In operation level, raw material information is fed into the inventory core system (Warehouse) in different formats/sources. EAI (tools such as Informatica) is the core component to ingest the incoming source data in a clean/expected layout. Rationalization is the process of extraction of the master data from the various systems of record of both internal and external systems. After processing, to produce the cleansed raw data using EAI and Rationalization in Warehouse, the BI layer takes the execution responsibilities. BI analyzes the enterprise's raw data from the various sources of the system.

Therefore, the lower operational layer of Data Architecture deals with the processing of inventory data from end to end, ranging from raw material to shipping the finished products. Thus, the operational layer cuts across the entire phase of the business.

The next tactical layer BPI is used to improve the existing business operation to accomplish significant improvement in production. In our use case, let's say the raw materials are sourced from various locations around the globe. In doing the various analysis methodologies, the BPI system can come up with an efficient way of sourcing the raw materials for the inventory. Of course, the existing raw data is essential for any prediction/analysis. Effective BPI generates promising results operational efficiency and customer focus, which in turn improves the productivity and profitability of the business.

By definition, enterprise data strategy is the comprehensive vision and actionable foundation for an organization's ability to harness data-related or data-dependent capability. To emphasize the importance of Data Strategy, let me share an interesting answer by Bill Gates of Microsoft. When he was asked a question--"What is the most important asset of your company?" he replied--"Data". In our use case, by doing Data Strategy of the inventory system, it drives the business to be a customer-centric data driven culture. In general, legacy systems produce data silos that will get in the way of understanding customers. This is a big challenge; without a Data Strategy, it is next to impossible for any inventory system. Due to the characteristic of relevancy, which is contextual to the organization, evolutionary, and expected to change on a regular basis, enterprise data strategy is essential to build the comprehensive strategies necessary to make a real difference for the organization.

Application Architecture

In general, software application is designed to meet an organizational need in reality mode. As the business model is quite common in a similar industry, it obviously expects the software application to build with the common architecture to satisfy the business requirements of an enterprise. As a result, Application Architecture is built in a generic way to create the business system, which is required to meet the specific requirements of the business.

By definition, Application Architecture specifies the leveraging technologies. Technologies are easily used to implement information systems, such as data, processes, and interfaces. On top of that, Application Architecture describes the details of the internal components and the way they interact to build the complete information system.

In terms of the engineering principle, Application Architecture exhibits the execution steps and methods in the model of the system blueprint into the reality of the leveraging enterprise.

Applications are generally categorized in the following listed types, along with their related characteristics. The categorization is based on the nature of the business process:

S.No.	Application processing type	Characteristics	Sample
1	Data	It is completely data-centric without explicit user manual intervention	Customer store Payroll application
2	Transaction	On the receipt of user requests, system-centric data is updated with the received information in a system database	E-commerce application Financial trade app
3	Event	This system is based on the receipt of the interested events from the system environment; it is not necessary to process non-interested data points	Traffic control system Real-time dashboard
4	Language	Users' interventions are specified in a formal language to be processed by the underlying system. It is mostly involved in system programming	Compilers and interpreter Command processor

Irrespective of the preceding types of application, Application Architecture is designed into the logical groupings of the software components. These logical layers help you differentiate between the different kinds of tasks performed by the components. In turn, the system is easier to support the design principle of reusability across the platform.

Earlier, I was so confused about using the terms Layers vs. Tier. Now, my understanding is that layer describes the logical groupings of the functionality/components in an application. However, tier describes the physical distribution of the functionality/components on the underlying hardware systems.

Each layer can be implemented as a large scale component running on a separate server. It is the most commonly used web-based architectural model in the industry. As a common practice, six layers are designed in the Application Architecture of the industry, which are as follows:

• End User Layer: This is an individual who uses the product after it is fully developed and marketed. This layer is around the usage pattern of the end user. As a result of rapid technology growth in recent times, the End User Layer is essential to build for desktop, web, mobile, pad, and so on.
• Presentation Layer: This contains the end user oriented functionality responsible for managing user interaction with the core system. In general, it consists of components that provide a common bridge between the end user and core business logic encapsulated in the business layer. Presentation Layer consists of UI components to render the user output and UI processor for local processing.
• Server Layer: This implements the core functionality of the target system by encapsulating the relevant business logic. In modern design, this layer consists of components, some of which may expose service interfaces that other callers can use. It is termed as the heart of the system.
• Access Layer: This layer is a bridge between the core business/server layer and the persisted store layer. It is designed using the best access pattern and practices of the enterprise architecture. It has two key components, namely the Data Access Component (DAC) and Service Gateway (SG). DAC allows programmers to have a uniform and comprehensive way of developing applications that can access almost any data store. The SG component encapsulates the low-level details of communicating with a service by means of service interfaces. Moreover, SG provides an ideal location to provide common features, such as asynchronous invocation, caching, and error handling.
• Persistence Layer: As the application data is persisted in this layer, it provides access to data hosted within the boundaries of the system and data exposed by other networked systems. By design, data is accessed through services in modern Application Architecture.
• External Layer: This layer is designed to expose the functionality of the application as the services to the external customer. API is the popular term in the industry, through which business services are exposed externally to earn the profit by sharing the best services.

In conclusion, Applications Architecture is the art and science of ensuring the suite of enterprise applications to create the composite architecture with the characteristics of scalability, reliability, availability, and manageability.

Technology Architecture

Technology/Infrastructure architecture principles are defined in collaboration with operational staff. It is the duty of the application architect to correct any wrong assumptions that the team might make with regard to enterprise infra architecture. Traditionally, it covers the servers, desktops, storage, network, and so on.

In the current distortive and emerging technology world, collaboration is the key for success. On connecting and cooperating with various groups, it is easy to adapt into the latest trends instead of reinventing the wheel again on our own. Technology Architecture is highly influenced by this principle.

On playing the enterprise architect role, my experience educated me to insist on a high degree of collaboration with other types of architects in the system. It is expected to have a closer working experience with a solution architect to roll out the implementation of the specific technology and platform as part of the role. In fact, architecture is not at all specifically associated with a particular release of the software. If so, then it is probably not considered architecture:

As depicted in the preceding image, Technology Architecture layers start from the Network layers of LAN, WAN, or Remote Access. On top of the Network layer, the Security principles are laid with Identity, Isolation, and Permission models. Storage layer is designed on top of Network and Security layers. Platform resides on top of the Storage layer and is the foundation for any type of software application, which reaches/touches the end customer. Take a look at the following model:

The Open System Interconnection (OSI) model is an interesting area in Technology Architecture. It defines a networking framework to implement protocols in seven layers. As the lower/deep technical details, control is passed from one layer to the next, starting from the Application Layer to the bottom most Physical Layer of bits. The communication passes over the channel to the next station and back up the hierarchy.

Technology Architecture deals with these various layers and its essentials.