Hi,

You're right: async url fetch doesn't seem to provide any way to
measure fetch time.

Then, I have a proposal: why don't you schedule a task per fetch, this
task will then do a regular synchronous url fetch for which measuring
will be easy.

So, your original servlet / task will schedule as many tasks as you
have fetches and wait via a loop of Thread.sleep() until all fetches
are done and their results (incl fetch time) written somewhere
(datastore / memcache) that the original servlet / task can access to
complete the work.

The only question though: can the application afford to launch those
tasks as they will incurr some overhead : some tenths of milliseconds
of cpu to schedule a task and run it + cpu time to write result to
cacche or ds in order to be shareable with originator.

If not acceptable, I hope it will at least give you some other ideas
to follow

regards

didier

On Jan 29, 9:50 am, Fabrizio Accatino <[email protected]> wrote:
> Hello,
>
> I'm playing with async urlfetch. My ispiration was Ikai 
> athttp://ikaisays.com/2010/06/29/using-asynchronous-urlfetch-on-java-ap...
>
> I run some request in parallel. All works fine. But now I'd like to get info
> about execution time of each request. The target URLs I call are different
> so the response time are very different.
> I read the documentation but HTTPResponse does not expose a "execution
> duration" or similar value.
>
> Any idea?
>
> Fabrizio

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine for Java" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-appengine-java?hl=en.

Reply via email to