top of page

Using Data Fabric to build a Modern Data Platform

Updated: 3 days ago

The concept of data fabric integrates and manages data from diverse sources, enabling efficient data access across organizations. It emphasizes democratizing data, ensuring its availability, and enabling efficient data exchange through contracts and virtualization.


Optimizing Data Management and Governance with Data Fabric


Shift in Mindset

The concept of data fabric involves the integration and management of data through flexible, reusable, augmented, and sometimes automated data integration processes, or by copying data into a preferred target database. This approach enables easy data access for both business users and data analysts. In the past, data ownership by business entities was common, with data management and governance seen as an implementation science. In contrast, while data mesh assigns data product ownership to business domains, data fabric creates an interconnected semantic layer by consolidating data from different sources.


Key Principles of Data Fabric


Democratize data

In a business setting, data fabric fuels exploration and creativity to develop products quickly. The democratization of data is more than just technological advancement; it represents a cultural shift involving people, procedures, and mindset changes. By adopting these values, companies can unleash the complete power of their data and foster innovation. Introducing a data catalog will define and reveal data and its attributes universally through a SEARCH function, accessible to all within the organization.


Ease availability of data

An internal marketplace serves as a strategic tool for democratizing and facilitating the exchange of data within a company. Through this marketplace, data is made accessible, creating a centralized repository of data assets for analysts to utilize. Nonetheless, for the marketplace to operate effectively, it cannot simply provide unrestricted access to all data within the system. It necessitates the implementation of robust data management controls, encompassing aspects such as privacy, security, authentication, encryption, entitlements, user access management, device management, and data rights management. These controls can be efficiently overseen by embedding them as metadata in a data dictionary.


Exchange data efficiently

Data contracts facilitate the structured and, if necessary, automated sharing and transmission of data to data consumers when combined with ETL/ELT or virtualization capabilities. These contracts represent official agreements between a data provider and a data consumer, outlining the organization, layout, properties, and design of the data. Data virtualization seamlessly combines and showcases data from multiple origins without physically relocating the data. These agreements establish principles and regulations for data exchange, storage, removal, or archiving, while also guaranteeing the data's reliability and quality, ensuring it is trustworthy for all involved parties.


Innovate through data products

Data products refer to applications or services that provide insights, predictions, or recommendations. By creating data products that address real-world issues or improve user experiences, you can drive innovation. Some instances include customer databases, tailored suggestions, fraud detection algorithms, and tools for optimizing the supply chain. Instead of the typical 21-day period needed to prepare data for a product, data provisioning for a data product can be completed in just four hours, leading to substantial cost reductions.



Data Management and Governance- Tejasvi Addagada

12 views0 comments

Comments


bottom of page