I have a large AlloyDB Omni database, but my on-premise machines only support volumes up to 14TB.
The main challenge is that a single events table accounts for at least 80% of the total database size, making it difficult to distribute the data efficiently.
What I've Considered So Far:
Citus for Sharding – I looked into Citus, but I couldn't find official support for it in AlloyDB Omni.
Partitioning & Logical Replication – While partitioning could help, a single master node would still contain all data, which doesn't solve the storage limitation.
Postgres_fdw + Manual Sharding – Another potential approach is foreign data wrappers (FDW) to distribute queries across multiple AlloyDB Omni instances.
What I'm Looking For:
A scalable, officially supported solution for sharding AlloyDB Omni.
A way to split large tables across multiple machines without exceeding 14TB per node.
Ideally, an auto-managed solution that avoids complex manual data distribution.
Does AlloyDB Omni support Citus for distributed storage? If not, what’s the best approach to handle this scenario?
Any insights or recommendations would be greatly appreciated!