My primary goal for 2018 Q1 (what I'm working on when I'm not doing the obligatory day to day DBA stuff) is to learn the basics of Cosmos DB, identify a business process that could leverage it, build a prototype, and then deliver a lunch and learn for the organization. There have already been some application developers talking about Cosmos DB, and we felt our DBA team needed to jog out ahead of user demand. The writing is on the wall, and the day will come when being a Microsoft centric DBA means (at least) SQL Server, Azure SQL, and Cosmos DB.
Cosmos DB is huge ocean for potential exploration; there are a handful of different data models (ex: DocumentDB, graph, key-value, column-family), and how the platform in general allocates, uses, and bills resources in terms of resource units (RU) is a new concept that the DBA needs to grasp. But what keep my learning focused is solving a specific business problem. We tend to retain what we learn easier when we apply our leaning to an application, and the reward for learning is greater. Rather than just aimlessly reading blogs and following step-by-step tutorials, I'll familiarize myself with at least one implementation path from end to end (at this point it's looking like DocumentDB), and I'll have an actual deliverable that can be shared with others. Maybe I'll blog about my experiences here.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho