Thank you, whoever ran that. It looks like Go doesn't do all the much better. The scheduler started up 3 instances for that. With the peak (when I happened to check it) at 25 QPS. I would be curious to hear what the real latency was.
Total number of instances Average QPS* Average Latency* Average Memory 3 total 12.022 18.5 ms 6.8 MBytes Instances QPS* Latency* Requests Errors Age Memory Logs Availability Shutdown 25.083 14.8 ms 9144 0 19:52:14 7.8 MBytes View Logs Dynamic 10.850 26.9 ms 815 0 0:01:02 7.1 MBytes View Logs Dynamic 0.133 14.8 ms 21 0 0:01:02 5.6 MBytes View Logs Dynamic This is under the free quota, though, but with no datastore ops, just serving a string from memory. On Jul 10, 2012, at 11:59 AM, Jeff Schnitzer wrote: > On Mon, Jul 9, 2012 at 8:05 PM, Richard <[email protected]> wrote: >> Ok, so I sat with Jeff in a G+ hangout for a while tonight, and we could not >> figure it out - other than my app is confusing the crap out of the scheduler >> which is then not sending requests to instances. > > Updated: This may not be a scheduler problem. It may be that we > can't wringing more than 10 requests/second out of Python27 > application instances. > > Given a request that does nothing but write a single entity to a > datastore (typically 50ms, but sometimes varies), what should we > expect throughput per Python27 instance to be? > > Jeff > > -- > 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. > -- 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.
