I doubt dispatcher is problem as forkjoinpool with default (and lesser settings due to tests required by ops) were tried. No other dispatches would make sense in our case.
As a sidenote grpc.io endpoint which does - route to akka cluster instance where state is via sharding actor (artery is used) - persistent actor modifies state (cassandra cluster) - route response back and respond to grpc call This endpoint has performance of ~6k req/sec (akka is cpu bound here) which (considering the amount of work) is "nice" (once again I see performance metrics in c++ so I'm biased, but rapid development in akka somewhat erases some issues for me) So I don't see where http endpoint could spend so much time if it has only ~15% more performance over grpc. Once again my main issue here is original vs current tests. Sadly I can find proof for both cases - akka http unacceptable/acceptable - online. One here, second on unrelated site. However nothing still explains performance degradation, except for my error. https://groups.google.com/forum/#!topic/akka-user/qhZlh0KBl2A https://www.techempower.com/benchmarks/ I did not study methodologies used to validate any of those benchmarks. Possibly akka team should do that so we have reference point. M. -- >>>>>>>>>> Read the docs: http://akka.io/docs/ >>>>>>>>>> Check the FAQ: >>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html >>>>>>>>>> Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups "Akka User List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
