Will do!

BJ "learned" me some things in a private thread on the subject.

The topic is not imminently critical because I've managed to create a
pretty good wrapper (around ServiceTracker) that will suite our needs.

I'll certainly let you know if I actually do anything as I feel there is
still room for it.

- Ray


On Mon, Jan 13, 2014 at 9:55 AM, Thomas Watson <[email protected]> wrote:

> Are you talking about synchronization in the ServiceTracker implementation
> itself?  If so I expect that to be identical across all frameworks since
> the ServiceTracker is an implementation provided by the osgi companion code
> directly, not the the various framework/service implementations.
>
> I seem to recall discussions in the past in this area but thought we
> avoided doing anything for compatibility reasons.  I suggest you open a bug
> and possibly start a branch if you have some implementation improvements to
> propose.
>
> Tom
>
>
>
> [image: Inactive hide details for Raymond Auge ---01/09/2014 10:17:44
> AM---Recently I've performed some JMH benchmarking on the usecase]Raymond
> Auge ---01/09/2014 10:17:44 AM---Recently I've performed some JMH
> benchmarking on the usecases for retireving collections of services
>
> From: Raymond Auge <[email protected]>
> To: OSGi Developer Mail List <[email protected]>,
> Date: 01/09/2014 10:17 AM
> Subject: [osgi-dev] heavy synchronization of ServiceTracker impls
> Sent by: [email protected]
> ------------------------------
>
>
>
> Recently I've performed some JMH benchmarking on the usecases for
> retireving collections of services.
>
> I discovered that raw usage of the ServiceTracker is in fact very slow for
> most cases due to heavy synchronization (both equinox and felix seem to
> have similar synchronization, although all my tests were actually against
> equinox).
>
> Has this ever been discussed or reviewed?
>
> It seems that since this is such a core function any improvement here
> would greatly affect/improve performance overall (not that it's really
> bad... except under significant concurrency).
>
> i.e. lock free impl.
>
> Thoughts?
>
> --
> *Raymond Augé* <http://www.liferay.com/web/raymond.auge/profile>
>  (@rotty3000)
> Senior Software Architect
> *Liferay, Inc.* <http://www.liferay.com/> (@Liferay)
> _______________________________________________
> OSGi Developer Mail List
> [email protected]
> https://mail.osgi.org/mailman/listinfo/osgi-dev
>
>
> _______________________________________________
> OSGi Developer Mail List
> [email protected]
> https://mail.osgi.org/mailman/listinfo/osgi-dev
>



-- 
*Raymond Augé* <http://www.liferay.com/web/raymond.auge/profile>
 (@rotty3000)
Senior Software Architect
*Liferay, Inc.* <http://www.liferay.com> (@Liferay)

<<graycol.gif>>

_______________________________________________
OSGi Developer Mail List
[email protected]
https://mail.osgi.org/mailman/listinfo/osgi-dev

Reply via email to