# When does a database become too complicated to operate? — retained record

> Retained reddit record from reddit:database.

Canonical: https://highsignal.app/data/records/83bfa17f0b426098

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:database)  published 2026-09-20 17:03 ingested 2026-09-21 02:38

[original source ↗](https://www.reddit.com/r/Database/comments/1wlmi58/when_does_a_database_become_too_complicated_to/)

## retained text



```
I've noticed that database design discussions often focus on whether a schema is normalized, whether queries are fast enough, or whether the right indexes exist. But in production, the harder problem sometimes seems to be understanding the system six months later.

At what point do things like excessive indexes, triggers, stored procedures, partitioning, caching layers, replication, and denormalized tables start creating more operational complexity than they're worth?

I'm curious how others decide when **database complexity is justified by a real requirement** versus when it's simply accumulated over time.
```



document fetched 2026-09-21 00:33 · canonical https://reddit.com/r/Database/comments/1wlmi58/when_does_a_database_become_too_complicated_to
