How Integration Architects Contribute to Enterprise Data Governance
Enterprise data architecture can be perceived as an abstract concept that refers to data quality, integration, reliability, accessibility and reporting. It is actually much more than that: enterprise data architecture links to master data management and business strategy.
Integration Architects (a.k.a. Enterprise Architects) play a key role in defining IT rules of engagement to enable and govern the enterprise data architecture, policies, business rules, quality standards, reporting, dashboarding, etc. across the enterprise—basically, everything that relates to data. This ranges from data creation, management, collection, exchange, etc. However, there is no or generic blueprint for enterprise data architecture; most of such knowledge is contextual and tailor-made for a given organization.
In this post, I expand on the meaning of ‘enterprise data architecture’, with the role of the Integration Architect in governing such subject matter in context of a given business strategy.
Integration Architects are by nature capable of high-level of abstraction and anticipation: they drive the design and conceptualization of the relevant business rules which relate to data management. This includes how people collaborate, how data flows between teams and functions, more specifically between different data sources, master and slave systems and platforms, ensuring that ‘single version of truth’ (SVoT) and ‘single source of truth’ (SSoT) principles are leveraged to minimize duplication and discrepancies, and maximize scalability and resilience.
“SSoT is about data input optimization (integration, input / output synchronization), while SVoT is really about business analytics and reporting optimization (consolidation, alignment).” (virtual+digital, 2016)
Enterprise data architecture
Data architecture links logical and physical data constructs, resources, models and structures. It includes platforms and their data models, interfaces, data extract-transform-load mechanisms, data back-ups and restore solutions, data transfer, conversion, exchange, etc. and the infrastructure supporting the relevant data processes.
Key elements of data architecture include data structures, unstructured data, data modelling, analytics, data mining, data sequencing, cloud and hybrid cloud storage and computing, virtual machines and containers, enterprise service bus / hubs, data lakes and warehouses. Integration Architects must choose the relevant elements when putting together business data management strategies.
Data hub: centralized seamless data sharing between enterprise platforms (SOA approach, rather than point-to-point integration).
Data lake: repository of ‘raw data’, structured and unstructured, that needs to be processed and analyzed to identify patterns and value-added behaviours (typically used in data science and machine learning).
Data warehouse: central repository of structured data from multiple sources across the enterprise; data to be collected and analyzed based on given analytics patterns (business analytics)
Enterprise data governance
Data governance is about choices and making decisions: it mostly concerns structured data and repeatable data patterns, which relate to business processes, control of data repositories / platforms.
In a nutshell, Integration Architects are key actors and technical leaders in delivering robust data governance, ensuring:
Business scalability and agility
Enterprise data vision definition
Business / IT alignment
Economies of scope and scale
Integration cost optimization
Reduced business and IT risks
Process and technology alignment through data
IT landscape blueprint
IT standard definition and governance
Data reusability and compliance
Design authority, linking into platform implementations such as PLM, ERP, MES, CRM, etc.