Dr. Philip Russom, an independent industry analyst and thought leader for data management, writes about getting greater business value from distributed data in an article on Denodo’s Data Management blog.
His article highlights the potential benefits and challenges of leveraging distributed data in organizations. It emphasizes the need to access and utilize the wealth of data distributed across various cloud and on-premise systems to unlock new opportunities and achieve business value. However, the main hurdle is that each system has its own data models, semantics, interfaces, and security restrictions, making it difficult to leverage and reuse the data effectively. As hybrid and multi-cloud data architectures become more prevalent, the distribution of data further complicates the situation.
To prepare for harnessing the value of distributed data, organizations are advised to establish a data management team with diverse skills and build a robust infrastructure for data management including tools, standards, and governance. The article suggests starting with small-scale solutions using data virtualization technology or creating a logical layer over existing data architectures. Over time, organizations can scale up their distributed data solutions to handle larger volumes from numerous sources, potentially adopting an enterprise-scale data fabric.
Addressing distributed data offers several benefits, such as creating a single view of the customer, enabling real-time or on-demand refreshes for performance management metrics, facilitating self-service data practices, and establishing a data abstraction layer as the heart of the solution. The abstraction layer simplifies access to distributed data, insulates users from its complexity, and ensures consistency in security, governance, and data standards.
However, there are challenges associated with distributed data, including hybrid architecture complexities, multi-cloud considerations with different providers, and the risk of creating silos in the cloud due to suboptimal data migration practices. Overcoming these challenges requires finding valuable enterprise data, defining business value, and respecting data requirements such as privacy regulations.
Dr. Russom’s article concludes by highlighting the need for multiple data strategies, such as data virtualization, data fabric, and logical data architectures, to effectively address distributed data and maximize business value. The ideal strategies combine high value with minimal technical effort.
Read the full article here.
