stolon VS zombodb

I wonder how is this compared to Stolon https://github.com/sorintlab/stolon?

Hi there! My name is Eric Ridge. I'm the developer behind ZomboDB (https://github.com/zombodb/zombodb). It's a rust-based system for creating indexes in Postgres backed by Elasticsearch.

You can read more about it on our GitHub repo, but basically it brings most of the power of elasticsearch’s searching and analytics abilities straight into Postgres.

There is an extension to integrate Elasticsearch with Postgres: https://github.com/zombodb/zombodb

Not affiliated, but for anyone looking to do searches on data stored primarily in Postgres via Elastic, ZomboDB is pretty slick.

ZomboDB is a Postgres extension that enables efficient full-text searching via the use of indexes backed by Elasticsearch. https://github.com/zombodb/zombodb#readme

If your search needs outgrow Postgres' native search engine, you can use Postgres search with an ElasticSearch backend, using Zombo

https://github.com/zombodb/zombodb

It basically gives you a new kind of index (create index .. using zombodb(..) ..)

Zombo does at least promise to handle "complex reindexing processes" for you (which IME can be very painful) but yeah, I assume you'd still have to deal with shard rebalancing, hardware issues, network failures or latency between postgres and elastic, etc etc.

The performance and cost implications of Zombo are more salient tradeoffs in my mind – if you want to index one of the main tables in your app, you'll have to wait for a network roundtrip and a multi-node write consensus on every update (~150ms or more[0]), you can't `CREATE INDEX CONCURRENTLY`, etc.

All that said, IMO the fact that Zombo exists makes it easier to pitch "hey lets just build search with postgres for now and if we ever need ES's features, we can easily port it to Zombo without rearchitecting our product".

[0] https://github.com/zombodb/zombodb/issues/640