alt Hacker NewsI was using TimescaleDB some very long time ago, things have changed quite a lot since (it's now even named differently).
In my current setup I was thinking on doing both: upgrading postgresql to timescaledb (to archive old data etc.), and to deploy ClickHouse in parallel. I'm still considering whether to go big on PeerDB to get ClickHouse mirror or just deploy it separately without additional fragility layer.
Would you not recommend using timescaledb at all? I definitely want to avoid alpha-quality software pain, since PostgreSQL is one of the most rock-solid parts of the stack at the moment.
Replies
I would just run both and decomission the old one when a) all data is migrated, b) old data is no longer relevant and can be archived
Worked on peerdb. If you're able to batch changes on your end & push to both postgres & clickhouse, do that. Only move to peerdb when you know you need cdc
In my (minor) experience Timescale works fine. The developer experience is good and it is very convenient to be able to JOIN against your hypertables. My only real complaints are operational (no logical replication, normal postgres update complaints), but man Clickhouse is really slick. I wrote some small reviews of the two in my submission history if you want a bit more detail.