Unifying Enterprise Data: How a Data Fabric Converges Application and Data Integration
Segregated Enterprise Data
Companies today face ongoing pressure from fast-changing customer expectations, innovative rivals, and new regulations. They understand that data is the key to quickly identifying and responding to a fluid business environment. Thus, many organizations have a growing ecosystem of applications and data sources that help them run their business and make decisions. With data in different locations and diverse formats, they need to share data between applications or prepare it for analytics through integration.
Integration tools and methods fall into two distinct categories: application integration and data integration. Application integration distributes data between different operational systems to improve efficiency. Data integration aggregates and harmonizes diverse data for analytical purposes. Unfortunately, this separation complicates things, making it hard to use data from one sphere in the other. For developers, this creates redundant work and data inconsistencies. For data consumers, it limits their self-serve options to those assets managed with data integration. They can’t access valuable operational data because it’s ensconced behind technical barriers they don’t know how to surmount.
Solution Approach
Organizations can bridge the gap between the operational and analytical data realms by using a data fabric that connects to enterprise application integration (EAI) tools. Data fabric is an architectural approach for analytics that integrates and simplifies data, making it more easily accessible for business users. On the other hand, EAI tools are designed to enable software engineers to synchronize data among operational applications in real-time or in small batches. Integrating EAI tools with a data fabric provides a comprehensive view of operational and analytical data and facilitates the bidirectional flow of data between them. This approach also provides a common user experience that serves both business and technical users.
Recommendations
To bridge the worlds of operational applications and analytics, consider the following steps:
> Identify and prioritize analytical use cases that require operational data. Evaluate analytics demand for operational data to determine the extent to which your organization can benefit from implementing EAI as a source for a data fabric. Interview data analysts, data scientists, and business leaders. Review backlogged requests for application data.
> Identify and prioritize use cases that require analytical data embedded in operational applications. Ask business leaders and operational workers how having analytical data points in their operational systems can streamline their work and improve their decisions.
> Evaluate current integration capabilities. Assess your existing data and application integration landscape to determine how ready your organization is to implement a combined EAI-and-data fabric solution.
> Prioritize data quality. Implement or enhance data quality monitoring to prevent misguided analytical conclusions.
> Educate and empower users. Provide training and resources to both technical and business users on the benefits of a unified data and application integration framework.
> Monitor and optimize costs. Keep a close watch on compute and storage costs associated with increased data access. Evaluate and prioritize analytical initiatives based on the cost of resource consumption versus the benefit to the organization.