Hello,
I have tried to get help tp one question like 1 year already. No help from the stackoverflow or Play Framework Google Group. Anyway, we have again performance issue in production. Our product is build on the top of Play Framework 2.1.4. Based on NewRelic report all the bottlenecks seems to be some where else than in our code. Could someone help me to understand the purpose of following three methods? I have already realised how these method calls are super slow in usage peak hours. Why is it so? Increasing parallelism-max value does not help much.
scala.concurrent.impl.CallbackRunnable.run()
scala.concurrent.impl.Future$PromiseCompletingRunnable.run()
org.jboss.netty.handler.codec.http.HttpRequestDecoder.unfoldAndFireMessageReceived()

application.conf
play {
akka {
event-handlers = ["akka.event.slf4j.Slf4jEventHandler"]
loglevel = WARNING
actor {
default-dispatcher = {
fork-join-executor {
parallelism-min = 350
parallelism-max = 350
}
}
exports = {
fork-join-executor {
parallelism-min = 10
parallelism-max = 10
}
}
}
}
}
Best regards,
Markku