Relational databases have "narrow regimes where they provide peak performance. [...] there are data-intensive tasks for which relational databases have been rejected: text indexing, media delivery".
Two research directions:
Programmer productivity a key challenge, declarative programming seems promising. Examples of potential:
Idea: Extend techniques from database, logic programming into new programming environments.
Challenge: extract structure and meaning from un-/semi-structured data.
Challenge: Effectively querying heterogeneous data.
"Tradeoff between functionality and operational costs: [2008]'s early cloud data service offen an API that is much mroe restricted than that of traditional database systems, with a minimalist query language and limited consistency guarantees".
Limited human intervention (no DBAs), high-variance workloads and infrastructure sharing increases importanve of "self-managing" research of last decade.
Current designs cannot scale to thousands of nodes. Unclear how storage will change (different implementation, storage semantics or both)
The term "metaverse" wasn't coined yet, but it seems similar:
"While [Virtual worlds like Second Life] began as interactive simulations for multiple users, they increasingly blur the distinctions with the real world, and suggest the potential for a more data-rich mixture. The term co-space is sometimes used to refer to a co-existing space for both virtual and physical worlds."
ad-hoc dblp analysis indicates that db research community doubled in size over the last decade. This places pressure on publications: