Closed kennedyoliveira closed 7 years ago
Hi. In such scenario, I would probably introduce Akka cluster sharding for the IssueRepository actor and add more Cassandra instances in its cluster. In such case, I would have an even distribution of load over the cluster of application instances (sharding region). You can take a look at Reactive Flows. As far as I remember, that example covers scaling too while my example is mostly focused on ES and CQRS.
Hello @BranislavLazic, thanks for the answers, i'll check the link you pointed, looks like a great example too! I'll check akka cluster sharding, i still didn't used this one. I have a question, maybe learning the cluster sharding would answer it, but i'll ask anyway, using sharding, the IssueRepositoryManager would still be needed?
Yes. To forward messages toward issue repository sharding region.
Thank you!
Hello!
Thank you for making this source code available, it's very useful, i'm studying akka and searching for patterns, best practices and ways of doing things, and i have a doubt based on this implementation.
How it's supposed to scale? Since it creates a new Actor for each issue, if a have billions of issues, how can i scale it?