In my blog post last month, I started looking at the concept of data fabrics to get an understanding around what it is all about. This month, I continue with the discussion, focussing on the advantages of data fabrics. The points I have outlined below are based on a very good article written by Lori Witzel, Director of Research for Analytics and Data Management at TIBCO and published on ITProPortal. I have added my own views here and based on my experience working in the data space.
To recap, the concept of data fabric was created to address the need for more data-driven insights while coping with the reality of the distributed nature of modern data architectures. Complicating this is that most organisations are dealing with data sources located on-premises and across hybrid and multi-cloud environments. For example, a company might be running both a CRM application and a modern data warehouse platform across two different cloud providers. A correctly implemented data fabric framework enables us to work across all these data environments.
However, it is not only about the technical connectivity and the integration of data flows, data access, and data storage. There is also an element of augmented data management and data governance that is required in such a cross-platform orchestration.
The challenge is that in the modern disparate, and often siloed organisation, a lot of de-duplication, verification, integration, and other data resolutions are required to get a complete and single source of the truth. Add to that the vast amount of ‘old’ data. While this might not operationally be required anymore, the company may need this data archived for trend analysis, historic reporting, or even as mandated by legislation.
The referenced article lists the four key advantages of using data fabrics as insights, innovation, information governance, and insured trustworthiness.
Insights
According to the author, data fabrics allow an organisation to treat its data like any other business component, something advocates have been crying out for years. Data fabrics not only allows the organisation to take advantage of more advanced insights, such as those derived through analytics and machine learning, but it also enables the data custodians to automate and accelerate data management. I would love to explore how this will work in a future post, so keep an eye out for that as well.
Innovation
The second point almost happens naturally. An organisation that leverages a data fabric approach can put their entire ‘data estate’ (as the article refers to it) to work. With so much more actionable insight being generated as a result, it is easier to transform the organisation through data-driven insights and attempt and adopt new levels of innovation. Of course, the organisation’s culture and agility must be able to support such an approach.
Information governance
This advantage sounds like music to my ears – because as our data environments become more complex, so does the data governance aspect too; coupled with, as the article states, an increasingly complex array of regulatory and compliance needs, not to mention more stringent requirements for privacy and security. The unified view provided through data fabric frameworks can simplify and streamline this complexity. I am eager to explore this in more detail too in subsequent posts – how exactly does a data governance forum utilise the facilities and features of the data fabrics framework to improve data governance over a faster flowing and more tightly integrated architecture?
Insurance
The name of this advantage was slightly strange for me as this point is about the trust in data – which of course is critical. I guess the author wanted to stay with the four ‘I’s theme. So, the data fabric provides improved trustworthiness in the data, and what can then be done with it. This flows directly from a more unified approach to data management. It will be interesting to see how this is physically implemented. As we all know, the downstream data quality can only be built on the basis laid in the source systems. Looking ahead, I am interested in examining each of these four concepts and how they will be technically implemented within an organisation.
