Delay in starting Java based Opendaylight controller

sonia verma soniaverma9727 at gmail.com
Thu Nov 7 00:52:44 PST 2013


Hi Andrew,

Yes the system is CPU-bound.
Also can you suggest me the JIT cross compiler for linux powerpc so as to
fasten up the process.
Is there any other way speeding up the process i.e without using any JIT
compiler?

Thanks



On Thu, Nov 7, 2013 at 2:12 PM, Andrew Haley <aph at redhat.com> wrote:

> On 10/31/2013 12:40 PM, sonia verma wrote:
>
> > I have ported Opendaylight controller  in one board having powerpc
> > architeture e500 kernel 3.8 supporting java:
> > java version "1.7.0_25"
> > OpenJDK Runtime Environment (IcedTea 2.3.12) (25b30-2.3.12)
> > OpenJDK Zero VM (build 23.7-b01, interpreted mode)
> >
> > When I run ./run.sh in order to start Opendaylight.it takes approximately
> > about 10 min for the controller to start properly so as to establish
> > connection with the OVS.
> > However when i run it on x-86,it is fine and is able to connect with the
> > OVS within no time.The java -version at x-86 is as follows..
> >
> > java -version
> > java version "1.7.0_25"
> > OpenJDK Runtime Environment (IcedTea 2.3.10)
> (7u25-2.3.10-1ubuntu0.12.04.2)
> > OpenJDK Server VM (build 23.7-b01, mixed mode)
> >
> > I noticed that the java mode at x-86 is mixed mode whereas at that of my
> > board is interpreted mode.
> > Can this be the reason for delay in starting?if yes suggest me how to
> > migrate from inerpreted mode to mixed mode .
> > Or there is some another reason associated with that.
>
> Zero is bound to be much slower than x86 because there is no JIT compiler.
> Having said that, ten minutes is an extraordinarily long time.  Is the
> process CPU-bound for that ten minutes?
>
> Andrew.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/zero-dev/attachments/20131107/2ed1983b/attachment.html 


More information about the zero-dev mailing list