Java in a docker container
kant kodali
kanth909 at gmail.com
Thu Dec 15 09:59:29 UTC 2016
Thats a very good response! I do think there are Brilliant minds that are
behind JVM and I think we should do everything possible for it to be #1. I
honestly think JLS itself had taught me way more than my 4 year degree from
a top school in US. Google has time and money so they can do anything. I
don't think oracle is doing much to promote Java or open source. I guess
they don't understand it.
On Thu, Dec 15, 2016 at 1:45 AM, Volker Simonis <volker.simonis at gmail.com>
wrote:
> On Wed, Dec 14, 2016 at 11:36 PM, kant kodali <kanth909 at gmail.com> wrote:
> > Hi Brian,
> >
> > If you think it is insufficiency of Docker then perhaps one may want to
> > look at how things are done in other Languages such as Go and so on? I
> just
> > don't why Java is targeted all the time when I hear from people saying
> "No
> > one puts Java in a container". My mind immediately thinks is it only Java
> > or any VM based language.
>
> I think part of the problem is that Java is one of the most
> sophisticated VMs (and unfortunately this sometimes translates into
> most most heavy-weight VMs :( It has two JIT compilers, several quite
> complex GC implementations which are by far better than the ones from
> other VM based languanges and a huge class library. This all adds up
> to a relatively big footprint and relatively slow startup.
>
> For example Java was newer designed for executing something like
> shell-scripts and the built-in ergonomics aren't designed to optimize
> the VM for such scenarios. They are better suited for web servers or
> heavy-weight applications like IDEs and they mostly assume they can
> grab most of the resources of the machine they are running on.
>
> I think these are the reasons why other scripting or VM-based
> languages run better "by default" in container enviroments - they
> simple need fewer resources by default. But I think Java can be tuned
> in the same way - we just have to do it :)
>
> Regards,
> Volker
>
> >
> > thanks,
> > kant
> >
> >
> > On Wed, Dec 14, 2016 at 8:48 AM, Brian Burkhalter <
> > brian.burkhalter at oracle.com> wrote:
> >
> >> There are also container-related issues in Java NIO [1]. I spent a
> >> not-insignificant amount of time on this a few months ago but a solution
> >> was elusive and we let it drop for JDK 9. I am not convinced that the
> >> fundamental problem is not due to an insufficiency of Docker itself.
> There
> >> are several issues linked from [1] among which the Docker-specific one
> is
> >> [2].
> >>
> >> Regards,
> >>
> >> Brian
> >>
> >> [1] http://mail.openjdk.java.net/pipermail/nio-dev/2016-
> >> October/003915.html
> >> [2] https://bugs.openjdk.java.net/browse/JDK-8165852
> >>
> >> On Dec 13, 2016, at 12:14 AM, Volker Simonis <volker.simonis at gmail.com>
> >> wrote:
> >>
> >> > […] and there's another Enhancement request open for this topic:
> >> >
> >> > 8146115: physical memory does not see Docker resource limits
> >> > https://bugs.openjdk.java.net/browse/JDK-8146115
> >> >
> >> > Please feel free to contribute any insights from your side to this
> >> discussions.
> >>
> >>
>
More information about the jdk9-dev
mailing list