Data Architecture Definition: Identifying the data needs of the enterprise (regardless of data structure) and designing and maintaining a master plan to meet those needs. Use a master picture to guide data integration, control data assets, and align data investments with business strategy.

Data architecture goals : (identify requirements; design structure; strategic preparation)

Identify data storage and processing needs.
Design a structure and plan to meet the enterprise's current and long-term data needs.
Strategically prepare organizations to rapidly evolve their products, services and data to exploit the business opportunities inherent in emerging technologies.

Architecture : The organized design of component elements to optimize the functionality, performance, feasibility, cost, and user experience of the entire structure or system.

The basic structure of a system: the components embodied in the architecture, their interrelationships, and the principles that govern their design and evolution. Carry out in different scopes and levels of the organization.

Architect Responsibilities: Use your own professional skills to clearly define and design content that is easily difficult to understand or confusing for non-professional architects so that it is easy to understand ( responsible for clearly defining things that are difficult to understand ).

Types of enterprise architecture:

business architecture
data architecture
Application architecture
Technology Architecture

Good architecture can help: Good enterprise architecture management helps organizations understand the current state of the system, accelerate the transition to the desired state, and achieve the goal of complying with regulations and improving efficiency.

The main goal of data architecture is to effectively manage data and the systems that store and use data.

The basic components of data architecture:

Data architecture results , including models, definitions, and data flows at different levels, which are often called components of data architecture
Data architecture activities to form, deploy, and achieve data architecture goals
Data architecture behaviors , including collaboration, mindset, and skills among the different roles that impact enterprise data architecture

Data architecture is the foundation of data management. Since most organizations have more data than an individual can understand, it is necessary to describe the organization's data at different levels of abstraction in order to better understand the data and help management make decisions.

Components of data architecture:

Description of current status
Definition of data requirements
Data integration guidelines
Data asset management specifications

The most detailed data architecture design document: It is a formal enterprise data model that contains data names, data attributes and metadata definitions, conceptual and logical entities, relationships, and business rules. The physical data model also belongs to the data architecture document, but the physical data model is the product of data modeling and design, not the product of the data architecture.

1.1 Business drivers

The goal of data architecture ( business drivers ) : to establish a smooth bridge between business strategy and technical implementation. Data architecture is part of the enterprise architecture.

Main responsibilities of data architecture :

Leverage the business advantages brought by emerging technologies to strategically help organizations rapidly transform products, services and data.
Convert business requirements into data and application requirements to ensure that valid data can be provided for business process processing.
Manage and deliver complex data and information across the enterprise.
Ensure business and IT technology are aligned.
Provide support for enterprise reform, transformation and improvement of adaptability.

1.2 Data architecture results and implementation

Key results of data architecture :

Data storage and processing needs.
Design a structure and plan that meets current and long-term data needs.

Data Architect: Architects seek to design an organization's data architecture in a way that brings value to the organization. This value is mainly reflected through appropriate technology application, effective operations, improved project efficiency and enhanced data application capabilities .

The main tasks of a data architect are:

Define the current state of the data.
Provides a standard business vocabulary for data and components.
Ensure data architecture is consistent with corporate strategy and business architecture.
Describe data strategy needs.
High-level data integration outline design.
Integrate enterprise data architecture blueprint.

Overall data architecture implementation :

Use data architecture components (master blueprints) to define data requirements, guide data integration, manage data assets, and ensure that data project investments are consistent with corporate strategy.
Work with, learn from, and influence stakeholders involved in improving business or IT system development.
Build an enterprise data language through data architecture and common data vocabulary.

1.3 Basic concepts

Architecture framework : The architecture of architecture. Ways of thinking and understanding architecture.

Enterprise architecture types: business architecture, data architecture, application architecture, technical architecture

Zachman framework:

basic concept
- The most famous enterprise architecture is the Zachman framework. The Zachman framework is an ontology, that is, a 6x6 matrix that constitutes a set of models that can completely describe an enterprise and the relationships between them. It does not define how to create models, it just shows which models should exist.

Zachman Framework—Inquiry Communication :
- What: Directory columns, the entities that build the schema.
- How: Process column, represents the activities performed.
- Where, distribution column, business location and technical location.
- Who: Responsibilities column, role and organization.
- Time When: Time column, representing intervals, events, cycles and schedules.
- Why: motivation column, table of goals, strategies and means
Zachman Framework—Redefining Transformations :
- 1) Executive perspective (business background): Define the business element catalog of different model scopes.
- 2) Business management perspective (business concept): Clarify the relationship between different business concepts involved in the defined business model by management.
- 3) Architect perspective (business logic): As a model designer, the architect refines the system requirements and designs the system logic model.
- 4) Engineer perspective (business entity): Engineers, as concrete model builders, optimize and implement physical models designed for specific applications within specific technology, personnel, cost and time constraints.
- 5) Technician perspective (component assembly): A technology-specific, context-free perspective is used to explain how technicians who configure the model use, assemble, and implement configuration components.
- 6) User perspective (operation category): actual functional examples used by participants. There is no model for this perspective.

Enterprise data architecture

Definition: Standard terminology and design that defines elements important to an organization. The design of enterprise data architecture includes business data description, such as data collection, storage, integration, movement and distribution.
details as follows:
- 1) Enterprise data model (data structure, data specifications).
- 2) Data flow design.

The concepts in enterprise data architecture are as follows:

Data : reports and analysis that need to be secured, integrated, stored, recorded, classified, shared, and ultimately delivered for use. During the process, the data may be: verified, enhanced, linked, authenticated, integrated, desensitized and used for analysis until the data is archived or purged.

Enterprise data model : It is an overall, enterprise-level, independently implemented conceptual or logical data model that provides enterprises with a common and consistent data view. Simplified abstraction. Includes: data entities (such as business concepts), relationships between data entities, key business rules and some key attributes.

Data flow design : Define the requirements and master blueprint between databases, applications, platforms and networks (components). Demonstrate the flow of data between business processes, different storage locations, business roles, and technology components.

Enterprise data model: Adopting industry-standard models can speed up the development of enterprise data models. As the needs of the enterprise change, the scope and content of each level in the enterprise data model will also expand, and it can be built in incremental and iterative ways at different levels. Entities in each enterprise data model should belong to only one subject area, but can be related to any other subject area. The enterprise conceptual data model is constructed by combining subject domain models and can be top-down or bottom-up.

The identification criteria for subject areas must be consistent throughout the enterprise model. Commonly used subject area identification criteria: use normalization rules to separate subject areas from system portfolios, form subject areas from data governance structures and data ownership (or organizations) based on top-level processes (business value chain) or based on business capabilities (enterprise architecture) . The subject area structure is usually most effective for data architecture work if it is formed using normalization rules. The normalization process will establish the main entities that host/make up each subject area.

Data flow : The data processing process that records the origin of data and is used to describe how data flows in business processes and systems. Where it comes from, where it is stored, and how it is transformed, data lineage analysis helps analyze and explain the data status at a certain point in the data flow.

Data flow mapping records the relationship between data and:

Application in business processes.
A data store or database within an environment.
Network segment (helps with secure mapping).
Business roles (describe which roles have responsibility for creating, updating, and deleting data).
Where local differences occur.

Data flow can be used to describe the mapping relationships of models at different levels: subject areas, business entities, and even mapping relationships at the attribute level. Presented as a two-dimensional matrix or data flow diagram.

2. Activities

Two ways to simplify your data and enterprise architecture :

Oriented to quality (consistent with tradition).
Be geared toward innovation (without being exhaustive).

2.1 Establish enterprise data architecture

Work included in establishing enterprise architecture : ( can be done in parallel or serially )

strategy. Choose a framework, develop a methodology, and develop a roadmap.
Communication and culture.
organize. Clarify responsibilities and accountabilities.
work method. Align with enterprise architecture.
result. Output in the overall roadmap.

Enterprise data architecture affects the scope boundaries of projects and system development:

Define project data requirements. Provide enterprises with data requirements for each project through data architecture
Review project data design. Conduct design reviews to ensure that conceptual, logical, and physical data models are consistent with the architecture and consistent with the organization's long-term strategy.
Determine the impact of data traceability. Ensure that the business rules of data flow in the application are consistent and traceable.
Data replication control. Replication is a common method that can improve application performance and facilitate data retrieval, but it may also lead to data inconsistency. Data architecture governance ensures adequate replication controls (methods and mechanisms) to achieve the required consistency (not all applications require the same stringency).
Implement data architecture standards. Develop and implement standards for the enterprise data architecture lifecycle. Standards can be expressed as principles, processes, guidelines and blueprints
Guide data technology and update decisions. Data Architecture works with Enterprise Architecture to manage data technology releases, patches, and data technology roadmap policies for each application.

Steps to establish enterprise architecture:

1 Assessment of existing data architecture specifications
2 Development Roadmap: Describes the 3-5 year development path. The roadmap should be guided by a data management maturity assessment
- The roadmap includes:
  - 1) High-level milestone events.
  - 2) Required resources.
  - 3) Cost assessment.
  - 4) Business capability workflow division.
- Development steps: The business data-driven roadmap can be developed from the most independent business capabilities, and then deal with the business capabilities with higher degree of interdependence.
3 Manage enterprise requirements in projects
- Steps to take: At the project level, the process of defining requirements through a data model begins with a review of business requirements. Typically, these requirements are specific to the project goals and have no impact on the business. The process should also include the development of term definitions and other activities to support the use of the data

Activities related to enterprise data architecture projects :

Define scope. Ensure that the scope and interfaces are consistent with the enterprise data model.
Understand business needs. Obtain data-related requirements such as entities, resources, availability, quality and pain points, and business value.
design. Form detailed target specifications.
Implementation (When to buy. When to reuse data. When to build.)

2.2 Integrate other enterprise architectures

Architecture activities are embedded into the project process in such a way that:

Waterfall way.
Iterative way.
Agile approach (DevOps)

3. Tools

Data architecture tools:

Data modeling tools.
Asset management software. (Used to manage data resource directories, describe their contents and track relationships between them)
Graphic design applications.

4. Method

4.1 Life cycle prediction

Objects of architectural design:

current.
deployment cycle.
strategy cycle.
Retired.
prioritized.
limited.
Emerging.
reviewed.

4.2 Icon usage guidelines

Icon usage guidelines:

Clear and consistent instructions.
All chart objects match the description.
Clear and consistent line direction.
Consistent crosshatch display method.
Consistent object properties.
Linear symmetry.

5. Implementation Guide

Data architecture includes components, activities, and behaviors.

Data architecture implementation work content:

Establish an enterprise data architecture team and conduct problem sessions.
Generate data schema version.
Form and establish data architecture working methods in development projects.
Increase organizational awareness of the value of data architecture efforts.

5.1 Readiness assessment and risk assessment

Readiness Assessment and Risk Assessment:

Lack of management support.
Evidence of success is lacking.
Lack of trust from managers.
Incorrect decisions by management.
cultural shock.
Lack of experienced project managers.
One-dimensional perspective.

5.2 Organization and culture

Organizational and cultural dependencies (acceptance of data framework depends on):

Acceptance of architectural approaches.
Confirm that data is a business asset of the organization and not just IT's job.
The ability to abandon a local data perspective and embrace an enterprise-wide data perspective.
Ability to integrate architectural deliverables into project implementation.
Standardize acceptance of data governance.
Based on the enterprise layout, not limited to the ability of project deliverables and IT solutions.

6. Data architecture governance

6.1 Data architecture governance activities

Data architecture governance activities:

Project supervision.
Manage architecture design, lifecycle and tooling.
Define standards.
Create data-related artifacts.

6.2 Metrics

Data architecture metrics:

1. Architecture standard acceptance rate: Measures the closeness of the project to the established data architecture and the degree of compliance with the project and enterprise architecture participation process
2. Realize the trend.
- 1) Use/reuse/replace/discard measurements.
- 2) Project execution efficiency measurement
3. Business value measurement indicators.
- 1) Business agility improvements.
- 2) Business quality.
- 3) Quality of business operations.
- 4) Improvement of business environment

DAMA-DMBOK2 key knowledge compilation CDGA/CDGP - Chapter 4 Data Architecture

1. Score distribution

2. Summary of key knowledge

1 Introduction