Centralized or Decentralized: Can We Answer This Question Once And For All?
The question is as old as the data warehouse. We are prone to centralize implementations of data platforms for purposes of governance and consistency. It is the pursuit of the holy grail for one version of the truth, proven, documented, and auditable. Yet the contemporary data analyst wants autonomy, exercising decentralized decision-making rights and the ability to enrich data, while ensuring compliance to local (e.g., country specific) regulatory requirements. Can we centralize or decentralize? Yes. Should we? Well, here it seems, the answer is, “it depends.” Hybrid environments prevail, balancing enduring and (often) conflicting demands.