On 21/09/2020 15:52, Mark Thomas wrote: <snip/>
> That doesn't > exclude, of course, the possibility of improving how Tomcat handles > traffic like this. Looks like Tomcat could prune the closed streams less aggressively. At the moment it waits until there are maxConcurrentStreams + 10% in the map and then: - removes all closed streams without children - removes all closed streams with children (re-prioritising as appropriate) - if there are still more than maxConcurrentStreams + 10% streams in the map, removes closed final streams (that are likely part of the priority tree) until the number of streams in the map is less than maxConcurrentStreams + 10% or there are no more closed streams to remove This means, with defaults and the load profile in this test, that the size of the map increases to ~110 and then drops to ~5, increases to ~110 and repeats. I'm currently thinking about different pruning strategies. The associated memory footprint is also part of my thinking. Mark --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
