Newsletters




Exploring The Key Pillars of a Modern Resilient Data Architecture

<< back Page 5 of 5

Take stock of skills, talents, and deficiencies. “Construct a strong, functional team,” advised Patel. To bring resilient data architectures to life, it’s important to facilitate closer collaboration and handoffs between data producers who build and operationalize data pipelines, and data consumers like data sci­entists and business analysts.”

Patel also suggested methodologies such as DataOps that will provide “a struc­tured approach to orchestration and automation of the data lifecycle that includes proper development, test­ing, deployment, and monitoring of data pipelines.” O’Connor rec­ommended closing this talent gap by “automating the more repeatable tasks in the data architecture. Data quality profiling and improvements, data enrichment, and data catalog­ing and governance are areas where good tools can be found to automate a portion of the work.”

Look at the whole picture. As mentioned earlier, data resiliency has been elevated beyond the bounds of IT to a business priority. Focus on “building capabilities instead of systems,” Rivero advised. “Instead of looking at business functions as separate, discrete components of the larger organization, consider each business function as a capability that contributes to its overall success while using common functions and leveraging shared data. Thus, design­ing for interoperability instead of hoping for integration is a good first step.” Additionally, Rivero said, it is important to understand how service delivery is supported by the various business processes and helps to con­nect the dots from data collection to actionable insights to services to outcomes.

THE ULTIMATE AIM

Ultimately, the aim of a resilient data architecture is “to deliver data that is of high integrity—accurate, consistent, and enriched with con­textual information,” said O’Con­nor. This comes out of support “for master data management, data inte­gration and transformation, data profiling and quality management, and straightforward processes for enrichment,” she added. “A resilient data architecture must have the pol­icies, procedures, and operational capabilities to ensure data privacy is guaranteed.

<< back Page 5 of 5

Sponsors