Re: [akka-user] Configurable parallelism/elastic groupBy

Thanks Endre. Ticket created https://github.com/akka/akka/issues/21020 Perhaps I mislead by using the word "elastic" in my subject. I just need "configurable". That said, why would I need consistent hashing? It seems that if I'm just preserving order within a channel by ensuring that all stream

Re: [akka-user] Configurable parallelism/elastic groupBy

On Fri, Jul 22, 2016 at 2:14 AM, Richard Rodseth wrote: > Or perhaps you are talking about something like > > > http://doc.akka.io/docs/akka/2.4.8/scala/stream/stream-cookbook.html#Balancing_jobs_to_a_fixed_pool_of_workers > Yes > > but using the Partition stage instead of

Re: [akka-user] Configurable parallelism/elastic groupBy

Or perhaps you are talking about something like http://doc.akka.io/docs/akka/2.4.8/scala/stream/stream-cookbook.html#Balancing_jobs_to_a_fixed_pool_of_workers but using the Partition stage instead of Balance

Re: [akka-user] Configurable parallelism/elastic groupBy

Thanks. That gives me some terms to Google, but any further pointers welcome. I see references to FlexiRoute being replaced by GraphStage. Presumably that's what I need to study next? On Thu, Jul 21, 2016 at 12:59 AM, Viktor Klang wrote: > A routing stage with consistent

Re: [akka-user] Configurable parallelism/elastic groupBy

A routing stage with consistent hashing and then a N-Way merge to insert into db? -- Cheers, √ On Jul 20, 2016 9:09 PM, "Richard Rodseth" wrote: > I'm sure I've asked this before in numerous ways, but it's still an issue > for me. > > I have an ETL stream that reads

[akka-user] Configurable parallelism/elastic groupBy

I'm sure I've asked this before in numerous ways, but it's still an issue for me. I have an ETL stream that reads per-channel data and writes it to a destination without backpressure. Within a channel, order of writes must be preserved. So I want parallelism between channels, but not within. If