Re: Comparison of three different installations of vasudevaservice.com

Hi Priyadarshan,

I have a routine that does this activity
on schedule each night. It reports quite
a bit of info, including elapsed time,
bytes, number of files, # errors. It is
like wget; no rendering, it isolates
page fetching.

I used it to last night in concert with
Zope's database monitor to get an idea
of the effect of the Zope debug-mode setting.
(about 14% decrease in time when turned off)
I left a description in
/opt/zope/vsa_3/log/admin.log.

If you have time for more testing,
it would be great to get an idea of
changing the number of Zope threads
from the default and other things.

The public Zope instance is now running
Plone 2.0 final.

Jagrata

  (Priyadarshan) wrote:

> Yesterday I spent a lot of time testing our current three different  
> installations of vasudevaservice.com:
> 
> 1
> the official, public one, through Plone rc5:
> http://www.vasudevaservice.com
> 
> 2
> the same Plone instance, but through Plone 2.0 (via zopedev):
> http://zopedev.vasudevaservice.com/pub/vasudevaservice.com
> 
> 3
> The same Plone site (via export/import), but on zopedyhani, with Zope  
> 2.6.2 and a pre-release Plone
> http://www.vasudevaserver.com:7091/vasudevaservice.com/
> 
> I found this tool to measure different things, among which total  
> filesize, number of items downloaded, download time, etc:
> 
> http://www.webpageanalyzer.com/
> 
> Open three tabs in firefox (or mozilla) with this urls, and see by  
> yourself the difference:
> 
> http://www.websiteoptimization.com/cgi-bin/wso/wso.pl?url=http%3A// 
> www.vasudevaservice.com
> 
> http://www.websiteoptimization.com/cgi-bin/wso/wso.pl?url=http%3A// 
> zopedev.vasudevaservice.com/pub/vasudevaservice.com
> 
> http://www.websiteoptimization.com/cgi-bin/wso/wso.pl?url=http%3A// 
> www.vasudevaserver.com:7091/vasudevaservice.com/
> (this one has a little bit more text, due to replication of news items,  
> for catalog issues)
> 
> Note that the final plone (2) has many less objects to download.
> 
> Also note that the largest amount of time is lag time:
> With 27 total objects for the pre-release home page, that computes to a  
> total lag time due to latency of 5.4 seconds!
> 
> For the final page, with 17 total objects, that computes to a total lag  
> time due to latency of 3.4 seconds.
> 
> I must say that the rendering time should be computed too. if you have  
> 27 items, that accounts for a little less than a second, that has to be  
> added to those numbers.
> 
> priyadarshan
> Vasudeva Server
> 
> 
> --------------------------------------------------------------------------
> This message is sent to you because you are subscribed to the mailing 
> list <VasudevaService-K9y3B4ipJAZZAIG4hdpdiNBPR1lH4CV8@public.gmane.org>.


--------------------------------------------------------------------------
This message is sent to you because you are subscribed to the mailing list <VasudevaService-K9y3B4ipJAZZAIG4hdpdiNBPR1lH4CV8@public.gmane.org>.