Back to Blog

What Are the Data Integration Architecture Key Components to Maximize Data Flow

Image of Matt White
Matt White

One of the primary goals of integration in smart buildings is providing a unified outlook that facilitates robust historical data analysis. Integration architecture must be designed to support the analysis of data flowing from multiple systems, and this data must be unadulterated, wide-ranging, obedient to a set of rules, reliable, and up-to-date. This requires standardization when the architecture is first designed, and again once when deployed. 

Understanding data integration architecture key components is essential to maximize data flows and harness the true power of big data while addressing data quality, privacy and security. 

The Need for Data Integration Architecture 

When it comes to the built environment, an automated building management system (BMS) benefits greatly from integration. Well-architected integration improves communications between the central BMS and a building’s various systems and subsystems, including the smart monitoring devices. Data integration architecture’s key components incorporate models that then synchronize processing in order to map and model these data flows. 

Within this integration, historical data needs to be accessible to accommodate analytics applications that help improve a building’s performance. Key aspects of this architecture allow data to be captured, extracted, and transformed so that it’s easily accessible once stored. Strong data integration architecture enables more accurate insights, allows you to respond to changes quickly, and optimizes scalability.

Data Integration Architecture’s Key Components

Developing strong data integration architecture entails utilizing  an independent data access layer (IDAL) to normalize and connect related data and then convey that via an API to various interfaces and platforms. By imposing order on what often looks like chaos, data integration architecture helps analytics software navigate a sea of complex data. 

Key components: 


check-long-1-1

 

Interoperability

Solutions must be compatible with the rest of the architecture by utilizing bidding systems that ensure the interests of building owners and other stakeholders rather than specific vendors.
check-long-1-1

Best practices

Best practices are followed to limit the design of serial networks that negatively affect troubleshooting and the ease of future maintenance; design should also utilize edge IP devices to bring logic and processing as close to the data source as possible so that adjustments can be made quickly.
check-long-1-1

Cloud computing

Cloud-based infrastructure that utilizes analytics with machine learning (ML) capabilities to continuously check system performance, enhance efficiency, produce opportunities for savings and track implementation by vendors.
check-long-1-1

Data ontology

Data ontology, along with standards for tagging and modeling, should be used to unify systems, optimize integrations and promote data use higher within the architecture.


Some of the barriers to successful integration include: 

  • Antiquated manufacturer specifications too often disregard network security and ownership on projects, which then don’t comply with the client’s IT standards or infrastructure
  • Specifications for products incorporated into networks should be neither manufacturer- nor vendor-centric
  • While open protocols within the data architecture support integration, many marketing claims for hardware and software don’t actually enable these open protocols optimally in practice, limiting integration efforts 

To overcome these challenges, data architecture should be integrated holistically. All systems within the network should be able to communicate with each other effectively. This necessitates knowing how data is managed, what communication protocols are used and which API to use, along with knowing where data is stored and how it’s managed.

Data Flow: Knowing Where the Data Goes

One of the key components of data integration architecture involves knowing how building data flows through your systems and where it will eventually live. Documenting data resources and mapping out data flows allows specific datasets to be more readily retrieved and used when needed.

Proper data mapping should follow these steps:

  1. Define the movement of data with tables and fields, formatting these once the data’s been moved; data integrations should also define how often data transfers should take place. 
  2. Map data by matching fields from sources to their destination fields. 
  3. Should a field require transformation, use a coded formula or rule. 
  4. Use sample data and a test system from the source to assess a data transfer first, making adjustments if necessary. 
  5. Deploy data integration or migration once data transformation has become workable.
  6. Maintain and update data map to continuously integrate data as new sources are added, when the data changes or when the data’s destination changes. 

Proper data management is essential to any modern organization. It allows businesses to initiate advanced analytics that generate insights, which then result in better decision-making. In built environments, this involves creating multiple layers of data architecture that optimize the performance of smart building management platforms and other data management tools. To achieve the best results, data architecture must also have specific standards for data collection, integration, transformation, and storage.

Data integration architecture should be collaborative in design to promote efficient processing. This helps optimize how efficiently data can be used across an organization, as well as improving its accuracy. Proper data mapping ensures that the right information goes to the appropriate areas and people. 

Using an MSI to Develop Data Integration Architecture

A master system integrator (MSI) should follow best practices throughout the integration process to ensure all networks and systems communicate with each other optimally. To achieve this, the MSI must instruct vendors to follow their direction rather than design their own portion of the architecture. 

Most vendors would prefer to silo their system using their own architecture and proprietary protocols to ensure customers will have to come to them for any future changes, service and updates. Some vendors advertise the use of protocols like BACnet to claim that they’re using open protocols. But they may apply only these “open” protocols to limited amounts of data.. While this might technically meet the specifications for their work, it restricts data flow and negates effective integration, locking customers out of the infrastructure they own. 

When networks are developed and owned by a vendor, customers are essentially held hostage, as credentials to control access to systems and data are held by the vendor. For systems that have not been kept up-to-date, this may cause additional problems. Equipment may no longer be supported by manufacturers, or devices run offline with overridden points. This means systems must run manually rather than automatically, negating much of the benefit smart building systems offer.   

An MSI can work with customers and vendors to prevent these issues and optimize the value of integration. That includes negotiating vendor service contracts that are already in place to ensure customers have full control and ownership. When integrating, this often involves migrating or upgrading any devices or software to facilitate delivery of data, mitigate risk, ensure reliability, and increase security. 

Buildings IOT: An MSI for Data Integration Architecture 

Buildings IOT’s expert MSI team streamlines communication between customers and vendors and resolves issues through unified solutions. Once building data is accessible, we fully assess network performance of underlying systems to ensure successful integration. Utilizing our in-house device qualification lab, we also validate which devices and manufacturers can be trusted within a network and determine how to properly configure hardware so that it works optimally for its intended purpose. 

Our designs guide MEP (mechanical, electrical and plumbing) teams to integrate smart buildings without creating additional risk. We ensure customers have complete control and ownership of their data integration architecture and verify vendors adhere to best practices while guiding the process through design, vendor procurement, and construction stages. 

Buildings IOT offers state-of-the-art products and services to support organizations' data integration architecture. Contact our team of experts to learn more about what we can do for you.

 

CONTACT US


Related Posts

Key Data Tagging Best Practices for Smart Buildings

Image of Jon Schoenfeld
Jon Schoenfeld

Smart buildings require data to operate at their most efficient. The more data gathered, the more...

Read more

Developing a Data Ontology Framework for an Intelligent Building Management System

Image of Jon Schoenfeld
Jon Schoenfeld

As more buildings become smart, there is a growing need for standardized semantic data models that...

Read more