Comment by coatue
7 days ago
[Joe, Hydra cofounder] Hey, that's really great - I love hearing that. Hydra is a columnar database with an integrated Postgres rowstore. Analytics aren't purely best on columnar: we've heard from users that their analytics workload would benefit from fast lookup on row tables too, not just scanning large tables. Our goal for Hydra is to enable realtime analytics on Postgres without requiring an external analytics database. This makes it possible to join the rowstore and columnstore data in Postgres with direct SQL. Other analytics databases typically rely on ETL pipelines to move data out of Postgres, which depending on your scale, can become expensive and introduce delay.
from what you wrote above, it seems like a great value add for greenfield projects.
we currently use aws aurora. how easy would it be to simply sql dump and load into hydra and how well would it serve as a drop in replacement?
Close to a drop-in replacement since Aurora bills itself as Postgres. Any data you load into Hydra will automatically be converted into the columnstore! we're happy to help out and feel free to DM me directly.