portals-pluto-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Martin Cooper" <mart...@apache.org>
Subject Re: Production quality
Date Wed, 07 Jan 2004 23:56:14 GMT

"Weaver, Scott" <Sweaver@rippe.com> wrote in message
news:BD961D4D4634D511BB990050DA1CE12C02967F99@winexch02.rippe.com...
> Mete,
>
> Funny you brought that up, David and I were just discussing that on IRC.
> David is going to work on pooling for PortletInvoker and, I think, the
> request and response objects.  We decided to implement it directly into
> Pluto instead of at the jetspeed level.  That way everyone can benefit
that
> is using Pluto.

Which interpretation of "Pluto" do you mean here?

* If you mean Pluto, the RI for JSR-168, then I strongly believe that Pluto
should not be including anything that isn't required for compliance with the
JSR itself.

* If you mean Pluto, the test portal, then I'm not sure I understand the
part about "everyone can benefit".

* Some other thing called Pluto? ;-)

--
Martin Cooper


>
> Regards,
> *================================*
> | Scott T Weaver                 |
> | <weaver@apache.org>            |
> | Apache Jetspeed Portal Project |
> | Apache Pluto Portlet Container |
> *================================*
>
> > -----Original Message-----
> > From: Mete Kural [mailto:metek@touchtonecorp.com]
> > Sent: Wednesday, January 07, 2004 9:20 AM
> > To: pluto-user@jakarta.apache.org
> > Subject: RE: Production quality
> >
> > Hello Scott,
> >
> > For instance let's say you have a very popular site with 100 concurrent
> > users. If there are 10 portlets on a page and you have to create 5-6
> > portlet objects (request, response, portletPreferences, etc) that adds
up
> > to thousands of objects to create. Could pooling provide performance
gains
> > when there is heavy usage like this?
> >
> > Thanks,
> > Mete
> >
> > ---------- Original Message ----------------------------------
> > From: "Weaver, Scott" <Sweaver@rippe.com>
> > Reply-To: pluto-user@jakarta.apache.org
> > Date:  Wed, 7 Jan 2004 13:01:54 -0500
> >
> > >Hi Mete,
> > >
> > >Pooling and caching is the domain of the portal and not the container.
I
> > am
> > >pretty sure the test portal in Pluto does not support pooling, as for
> > >caching I am not 100% sure.
> > >
> > >I really don't think pooling of portlets is needed.  The spec indicates
> > >there should be only one instance of a specific portlet's class per vm,
> > just
> > >like servlets.  Maybe if you can give an example of were pooling would
> > >increase performance, please do so as we may have missed something.
> > >
> > >
> > >Regards,
> > >*================================*
> > >| Scott T Weaver                 |
> > >| <weaver@apache.org>            |
> > >| Apache Jetspeed Portal Project |
> > >| Apache Pluto Portlet Container |
> > >*================================*
> > >
> > >> -----Original Message-----
> > >> From: Mete Kural [mailto:metek@touchtonecorp.com]
> > >> Sent: Wednesday, January 07, 2004 4:45 AM
> > >> To: pluto-user@jakarta.apache.org
> > >> Subject: Production quality
> > >>
> > >> Hello All,
> > >>
> > >> I'm wondering if IBM intended Pluto to be a production quality
portlet
> > >> container or just to test JSR-168 compatibility. As far as I
understand
> > >> there is no pooling of portlet objects and caching done in Pluto. Do
> > you
> > >> know if there are plans to add these features to Pluto?
> > >>
> > >> Thanks,
> > >> Mete
> > >
>




Mime
View raw message