There are multiple levels of looking at performance. Tool I wrote analyzes
server side performance since that was the main goal of project however
you could adapt it to analyze any other type of response times ie. there
is a parser that takes URLs and how long it took the web server to serve
that request. User perceived performance is much harder to gauge since it
depends on a number of factors not under your control such as CPU speed,
rendering engine, spyware etc. slowing things down etc. You should
obviously get as many metrics and evaluate how you can speed things up but
server side performance ie. time it takes to generate the HTML is one you
have most control over.
As far as client perceived response time there is also Boomerang
https://github.com/yahoo/boomerang
Vladimir
On Sat, 12 Mar 2011, Archana N wrote:
Thank you so much. This looks really interesting. I am working on a project to
monitor the performance of lighttpd servers in a
cloud environment(for resource provisioning purposes) . Also, in the blog, I was not
clear by the term "response time". Is it the
response time as perceived by the client or is it the request processing time
of the server. I had seen many tools such as Piwik
which would measure the client percieved response time of the web pages, but
they have the disadvantage of having java-script to
be enabled at the client's for this purpose. Is the tool you had developed
similar to Piwik or is it measuring the server side
processing time.
-Archana
On Sat, Mar 12, 2011 at 9:46 PM, Vladimir Vuksan <vli...@veus.hr> wrote:
You could use Ganglia but depending on number of unique URLs this may not
be such a great idea. I did something
similar at a previous job to evaluate page response times (aggregated on
hourly basis). You can find it here
https://github.com/vvuksan/pagetime-analyzer
I blogged about it here
http://vuksan.com/blog/2010/07/15/analyzing-your-web-page-response-times
Vladimir
On Sat, 12 Mar 2011, Archana N wrote:
Hi,
I was also thinking of having metrics for counting the number of access
of a particular string, but the problem
is I am working
with an application which has many directories (similar to wordpress) and
there are a lot of directories which
get different
amount of hits by the users and I would like to track which ones are
mostly hit. I will also try to think about
it. I was not
sure if Ganglia provided the aggregation, but now I got the answer to
that :) .Do let me know if you also come
up with ideas.
On Fri, Mar 11, 2011 at 6:34 PM, Rick Cobb <rick_c...@ieee.org> wrote:
Ganglia doesn't have a model for aggregating string-valued metrics.
On the other hand, you can get a long way by having your
metrics-gathering modules post a count *per string* (e.g.,
"www.yahoo.com:hits", 15); you'll have a ton, though, so you may want
to use some sort of naming prefix to help organize them.
If you have ideas about how you'd like to see them aggregated, I for
one would love to hear them; it's a fun problem to try to solve.
-- ReC
On Fri, Mar 11, 2011 at 9:13 AM, Archana N <dreamgirl...@gmail.com> wrote:
> Hi,
> I am using Ganglia for monitoring lighttpd server statistics in a cloud
> environment. In my case I would have string metrics such as the pages
which
> are frequently accessed etc. I understand that if there are many
clusters,
> then Ganglia aggregates the information at the grid level. This is
possible
> for the numerical metrics. However, I would like to know how this would
work
> for the string metrics.
> -Archana
>
>
------------------------------------------------------------------------------
> Colocation vs. Managed Hosting
> A question and answer guide to determining the best fit
> for your organization - today and in the future.
> http://p.sf.net/sfu/internap-sfd2d
> _______________________________________________
> Ganglia-developers mailing list
> Ganglia-developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ganglia-developers
>
>
------------------------------------------------------------------------------
Colocation vs. Managed Hosting
A question and answer guide to determining the best fit
for your organization - today and in the future.
http://p.sf.net/sfu/internap-sfd2d
_______________________________________________
Ganglia-developers mailing list
Ganglia-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-developers
