what are the differences between Timescale and Citus? When do I want to use which? AFAIK Citus scales multi-nodes long time already, your solution of hypertables reminds me of Citus distributed tables... Citus can do rebalancing long time as well... And much more. Would love to hear what you think makes you different.
Citus is a great product that primarily focused on multi-tenant SaaS analytics, where you typically shard your queries on the tenant_id, but didn't focus on time-series. So, their current documented solution for time-series uses other extensions (pg_partman), doesn't provide for compression, and all the other essential tools to make massive amounts of time-series data useful, which TimescaleDB is hyper-focused on solving.