A set of generic methods and rules to design a sturdy, cost-efficient, and scalable information mannequin in your post-modern information stack.
Over the previous few years, because the Fashionable Information Stack (MDS) launched new patterns and requirements for shifting, reworking, and interacting with information, dimensional information modeling steadily grew to become a relic of the previous. As an alternative, information groups relied on One-Large-Tables (OBT) and stacking layer upon layer of dbt fashions to deal with new use instances. Nonetheless, these approaches led to unlucky conditions during which information groups grew to become a value middle with unscalable processes. So, as we enter a “post-modern” information stack period, outlined by the pursuit to cut back prices, tidy up information platforms, and restrict mannequin sprawl, information modeling is witnessing a resurrection.
This transition places information groups in entrance of a dilemma: ought to we revert again to strict information modeling approaches that had been outlined a long time in the past for a very totally different information ecosystem, or can we introduce new rules which might be outlined primarily based on at the moment’s expertise and enterprise issues?
I consider that, for many firms, the appropriate reply lies someplace within the center. On this article, I’ll focus on a set of information modeling requirements to maneuver away from…