Relational database management system guide: RDBMS still on top
A comprehensive collection of articles, videos and more, hand-picked by our editors
The proliferation of multiple non-relational databases is transforming the data management landscape. Instead of having to force structures onto their data, organisations can now choose NoSQL database architectures that fit their emerging data needs, as well as combining these new technologies with conventional relational databases to drive new value from their information.
Until recently, data’s potential as a source of rich business insight has been limited by the structures that have been imposed upon it. Without access to the new database technologies now available, standard back-end design practice has been to force data into rigid architectures (regardless of variations in the structure of the actual data).
Inherently inflexible, these legacy architectures have prevented organisations from developing new use cases for the exploitation of structured and unstructured information.
The ongoing proliferation of non-relational database architectures marks a watershed in data management. What is emerging is a new world of horizontally scaling, unstructured databases that are better at solving some problems, along with traditional relational databases that remain relevant for others.
Technology has evolved to the extent that organisations need no longer be constrained by a lack of choice in database architectures. As front-runners have moved to identify the database options that match their specific data needs, we saw three key changes becoming increasingly prevalent during 2012:
- A rebalancing of the database landscape, as data architects began to embrace the fact that their architecture and design toolkit has evolved from being relational database-centric to also including a varied and maturing set of non-relational options (NoSQL database systems).
- The increasing pervasiveness of hybrid data ecosystems powered by disruptive technologies and techniques (such as the Apache Hadoop software framework for cost-effective processing of data at extreme scale).
- The emergence of more responsive data management ecosystems to provide the flexibility needed to undertake prototyping-enabled delivery (test-prove-industrialise) at lower cost and at scale.
From now on, savvy analytical leaders will be seeking to crystallise the use cases to which platforms are best suited. Instead of becoming overly focused on the availability of new technologies, they will identify the “sweet spots” where relational and non-relational databases can be combined to create value for information above and beyond its original purpose.
By taking advantage of the new world of choice in data architectures, more organisations will be equipped to identify and exploit breakthrough opportunities for data monetisation.
Just as communications operators have created valuable B2B revenue streams from the wealth of customer data at their disposal, so better usage of their existing data will empower other companies to build potent new business models.
For more on NoSQL database architectures and business value
What is NoSQL (Not Only SQL)?
Shawn Rogers podcast: data scientists can add business value, but not for everyone
Implementing a rethink of how data is stored, processed and enriched means re-evaluating the traditional world of data management. Until now, data has been viewed as a structured asset and a cost centre that must be maintained.
The availability of new database architectures means that this mindset will change forever. Data management in a services-led world will require IT leaders to think about how the business can most easily take advantage of the data they have and the data they may previously have been unable to harness.
Agile data services architecture
As more architecture options become available, data lifecycles will shrink and become more agile. Rather than seeking to “over control” data, approaches to data management will become much less rigid. One key aim will be to open up new possibilities by encouraging and facilitating data sharing. Amazon stands out as a pioneer in this field. By building a service-oriented platform with an agile data services architecture, the company has been able to offer new services around cloud storage and data management – as well as giving itself the flexibility needed to cope with future demand for as yet unknown services.
Unprecedented accessibility to non-relational databases is reinvigorating the role of conventional architectures and “traditional” data management disciplines. From now on, analytics leaders will increasingly move to adopt hybrid architectures that combine the best of both worlds to leverage fresh new insights from the surging volumes of structured and unstructured information that are now the norm. In summary, there has never been a more exciting time to be a data management professional.
About the authors