Select actor based on persistence id

I think my previous DDD / CQRS / ES mental model is preventing me from seeing the ‘AKKA’ light…

In my previous work, command handlers for updating aggregate roots typically take up the form:

var  aggregate = repository.load(id);

repository.load(id) will instance the AR, and apply events / snapshots before returning a rehydrated AR.
doBlah will cause the AR to emit an event which will both update it’s state, and be appended to the journal.

I’m looking to acheive the same with Akka(.Net). I see that persistence actors have a PersistentId but I don’t see how I can use it to get an actor reference so I can tell(DoBlah).

Usually, if there are no questions in the community about a thing, that means I’m asking for the wrong thing? Am I?


Having done some further research, I’ve found this blog post that presents an aggregator. Is this the idiomatic approach? I fail to see how clustering / Sharding would help if I still have to maintain a list of 1 million notes in this one aggregation?

Sharding will locate the Aggregate in the Cluster, and give you a reference. In akka classic the id of the Aggregate came from the command and in akka typed first you get the reference with the id and then send the command. Regarding the million notes: will affect the size of your actor nothing more.