MongoDB and similar.
WHAT THEY STORE
Documents, typically structured as nested records.
WHAT THAT PROVIDES
Flexible structure without a fixed schema Nested data retrieved in one read
WHAT IT COSTS
No enforced structure, so consistency is the application's problem.
WHY THAT MATTERS
Documents drift, and old ones lack fields new code expects.
WHAT TO DO ABOUT IT
Apply schema validation, which these databases support.
WHEN THEY SUIT
Genuinely varied structures Data naturally nested and read as a whole Rapid change during early development
WHEN THEY DO NOT
Data with clear relationships Anything requiring transactions across records, historically Reporting across many collections
WHAT INDEXING STILL REQUIRES
The same attention as any database.
WHY THAT SURPRISES PEOPLE
Flexible structure does not make queries fast.
WHAT TO MONITOR
Slow operations Index usage Memory against data size
WHAT TO SECURE
Authentication, which older versions did not require by default.
WHY THAT MATTERS
A great many such databases were exposed and stolen because of it.
WHAT TO CHECK ON ANY INSTALLATION
That it requires authentication, and is not reachable publicly.
