Has anyone had any success with YugaByte DB and Ecto?
a Google Spanner inspired Postgres derivative for planet scale horizontal db
1 Like
- It doesn’t seem like “derivative”, more like other DB with compatible API.
- Do anybody really need something like that? In 99% of use cases PostgreSQL + 1/2 hot replicas should do the job with much less troubles. And when you want to do more, then you always can use something like CitusDB that allows you to use some of the PostgreSQL niceties (plugins) without problems.
2 Likes
as it claims to be postgres compatible, maybe postgrex just works? https://github.com/elixir-ecto/postgrex
3 Likes
The biggest YugaByte advantage is that application can connect to any node in a cluster and get the same view on data as far as I’m concerned. And while PostgreSQL might be sufficient performance-wise I’v heard that master-master replication is a pain. To my understanding YugaByte solves that with a grace.
Ecto by itself doesn’t support that use case natively. Throwing in HAProxy or something similar would be required.
Again, familiar with YugaByte quite superficially.
1 Like
