We've added PSML into the database, Im doing the same for registries
now.
Goal is clustering Jetspeed
The aggregation engine is good, I did some performance tuning on it
recently.
I'd like to take a few days sometime and try to tune the engine.
Areas of improvement:
-- payload is always dependent on the slowest portlet.
-- there is no timeout on a portlet to generate its content
-- all portlets content is aggregated on the same thread.
-- caching of portlet container elements
We have caching built into to most of the major features.
IMO the logging is excessive, we should do something like
If (loggingON)
log("big long string constructor" + blah)
Instead of
Log(""big long string constructor" + blah)
Don't know how much that will improve speed, have to test it
Jetspeed logs the basic aggregation cycle to the log file
Look for things like:
[Fri Mar 22 11:08:40 PST 2002] -- DEBUG -- PortletFactory.getPortlet():
took a total of 0 millisecond(s) ->
-1653569551|template-hello|action-portlets.HelloAction|text-Hello World
in Velocity
> -----Original Message-----
> From: Saddest OfAllKeys [mailto:[EMAIL PROTECTED]]
> Sent: Friday, March 22, 2002 11:49 AM
> To: [EMAIL PROTECTED]
> Subject: scaling
>
>
>
> Is Jetspeed being architected so that it can scale?
>
> Does anyone have an experience or anecdotal evidence
> regarding load testing of a Jetspeed app?
>
> I'm blown away by Jetspeed, but my bosses are going to
> have serious questions before I can use it for my
> project.
>
> Mike
> [EMAIL PROTECTED]
>
> ps. I know this is a dev list versus a user list but
> I'd like to hear from a real hands-on guru, hence my message.
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Movies - coverage of the 74th Academy AwardsR
http://movies.yahoo.com/
--
To unsubscribe, e-mail:
<mailto:[EMAIL PROTECTED]>
For additional commands, e-mail:
<mailto:[EMAIL PROTECTED]>
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
