If you adapt dynamic sharding to map to ES shards instead of ES indexes, that will not fit my use case, because I still need to put some documents in ARCHIVE and others - in LIVE.

I see. Well then, I fear our biggest problem will be how to name this, given we already have a feature called "dynamic sharding" that doesn't really do "sharding" in Elasticsearch terms... We'll think about this.

I am implementing this ES integration in parallel with you, and it is not live yet. I am also searching for a way to extent Hibernate Search 5.6.1.Final to support _routing (even in a hardcoded way) but I cannot find an extension point for that.

Out of curiosity, even if it's not working yet, is this code available publicly?
By the way, if you want to work on it, feel free to send a PR adding this "_routing" support directly to our implementation (master branch, which is 5.8.0-SNAPSHOT). We'd be glad to merge it if it doesn't impair other features (though we may discuss the public-facing stuff), and it may be easier to implement since you'll be able to access internal engine code.