What to do when my data is based on time-series?
PostgreSQL can be used for such data, however there is an extension on top of it that has some benefits.
TimescaleDB can be used as an alternative for time-series data as it is based on PostgreSQL and uses the same syntax, while being optimized for this type of data. It can achieve 10 to 100x faster queries depending on the commit type you choose.
For the fastest commits, we recommend batch committing as otherwise you hit the limit of your system quite fast.
Furthermore, partitioning on a logical time-series base can also be useful as this type of data can grow quickly out of hand. When partitioning, it is always possible to delete older, unneeded data without worries.
The only downside of using TimescaleDB is the licensing. They use a dual license and some parts of it are not open source.
For further reading about TimescaleDB, head to https://timescale.com