We are seeing this problem too, our regular instance are 6-10 and they
were at 300-400

Our code is now disabled, it was using memcache, but we have no fix
for it so far, no change in code or traffic.

See this thread.
http://groups.google.com/group/google-appengine/browse_thread/thread/29515fdc51f80b32#

On Dec 8, 11:13 am, Rishi Arora <[email protected]> wrote:
> I'm seeing this too, but for me this has turned out to be a good thing.  I
> have my Max-Idle-Instances fixed at one.  But I thought my issue was more
> because I lowered my Max-Pending-Latency from 5 seconds down to 1 second,
> and then further down to 50ms.  In terms of impact on resource-usage, I
> have only seen my overall instance-hours used per-day go from ~27 (under
> the free amount of 28), to about 30.  To me, a cost increase of $0.16 is a
> small price to pay for a really low max-pending-latency of 50ms.  The user
> experience has been a lot better since then.
>
> For your case, keep in mind that QPS is only averaged over the last minute
> - so it may not be a good indication for overall load demanded of your
> instances.  I do feel that there was a change in GAE's instance scheduler,
> but I would recommend tweaking the scheduler settings  - more importantly,
> you may need to lower your Max-Idle-Instances, so that an increase in
> active-instance-count (in response to spike in incoming requests) doesn't
> have a huge impact to your cost.  Also, correspondingly, if you're reducing
> max-idle-instances, you may also need to reduce your max-pending-latency so
> that instances are spawned sooner to serve your incoming load.  Do you have
> a sense for how long your instance start up time is?
>
>
>
>
>
>
>
> On Thu, Dec 8, 2011 at 10:33 AM, Clément <[email protected]> wrote:
> > I'm experiencing a very high and unusual instance count on one of my apps
> > since this morning.
>
> > I have about 10 instances usually, but today it's stuck between 40 and 50.
>
> > There is no code change since yesterday, and the load is about the same as
> > every day.
>
> > The average QPS is extremely low (below 0.1), and the latency is also
> > quite low (below 500ms), so it doesn't make any sense.
>
> > I've almost exhausted my quota for today because of this issue.
>
> > Has something changed in the instance scheduler since yesteday ?
>
> > --
> > 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/-/DgFyJxrcopQJ.
> > 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