Hey Dave,
  So do you see pending latencies in the request log headers?

  What if you further increase the work done within each task so that
it takes, for example, 5 minutes per task?

  Have you considered using pull-queues since you're running all this
on backends?


Robert




On Tue, Jan 31, 2012 at 16:58, Dave Loomer <[email protected]> wrote:
> Hi Robert,
>
> When targeting backend "thebackend", this was done merely by specifying
> target=thebackend when creating the Task object. And then in backends.yaml I
> modify the instances parameter for the backend in question for each of my
> various tests.
>
> I'm not really too concerned about failfast, so I'll skip that one for the
> moment ...
>
> I haven't played with max concurrent requests, but in the test where I used
> a frontend, I did not see any additional instances spin up (and my site is
> very low-use, so there was probably a max of 1-2 instances to start with).
>
> I see the same length of delay (~20 seconds) whether I'm testing with my
> rapid-fire test handler, or running in production with tasks that spend 1+
> minutes doing actual work (lots of RPCs- datastore, URLfetch, memcache).
>
> Hope this helps.
> Dave
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google App Engine" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/google-appengine/-/xF79RpgjpAYJ.
>
> 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?hl=en.

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" 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?hl=en.

Reply via email to