[concurrency-interest] ForkJoinPool for async http calls?

Henrik Johansson dahankzter at gmail.com
Thu May 10 11:42:32 EDT 2012


Correct me if I am wrong here but asynchronous IO does not perform better
than it's synchrous counterpart. In order to get a performance increase you
have to think about performance in terms of throughput and scalability. If
that's the case then asynchronous IO and Play2 is the way to go. Play2 has
lots of other goodies too though...

The experts present in this mailing list can probably tell me otherwise if
I am wrong ;-)

/ Henrik
On May 10, 2012 5:31 PM, "Christian Essl" <christianessl at googlemail.com>
wrote:

> First thanks for this great work and its really an honor to me that
> such great coders like you answer my questions.
>
> The main reason I asked the original question is not that much about
> raw performance but more about code-simplicity in the sense to code
> like in a traditional blocking style but still have better performance
> and less blocking (not necessarily none).
>
> So I hoped to get a somehow sulution with my original attempt.
>
> Regarding CountedCompleter an example usage for IO would certainly
> help me, because I don't know where I should plugin the
> completion-callback for async-IO or alternatively do the blocking IO.
>
>
> Thanks
> Christian
>
> On Thu, May 10, 2012 at 2:57 PM, Doug Lea <dl at cs.oswego.edu> wrote:
> > On 05/10/12 08:12, Christian Essl wrote:
> >>
> >> As said I like the akka-framework, still it would be very nice if
> >> anybody could answer my original question whether ForkJoinTasks can
> >> give a sort of non-blocking future also suitable for async IO, because
> >> I am stil not sure.
> >
> >
> > The jsr166y/jdk8 CountedCompleter and related updates are specifically
> > aimed to better support IO-bound and other non-computationally-bound
> > tasks. The javadocs and examples probably don't yet do a good enough
> > job in showing how. (Partly because IO examples are always too bulky
> > to make for simple understandable examples; suggestions would be
> welcome.)
> >
> > The syntax and constructions for using them are not as nice as in
> > akka/Scala.
> > On the other hand the ability to effectively AND completion triggers
> > via counts (not found yet in Scala versions) is handy.
> >
> > -Doug
> >
> >
> > _______________________________________________
> > Concurrency-interest mailing list
> > Concurrency-interest at cs.oswego.edu
> > http://cs.oswego.edu/mailman/listinfo/concurrency-interest
> _______________________________________________
> Concurrency-interest mailing list
> Concurrency-interest at cs.oswego.edu
> http://cs.oswego.edu/mailman/listinfo/concurrency-interest
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.oswego.edu/pipermail/concurrency-interest/attachments/20120510/a0e9ebd0/attachment-0001.html>


More information about the Concurrency-interest mailing list