Update: We've made a production change that has resulted in more
predictable datastore behavior, reducing the number of latency spikes
that have been occurring. We'll be monitoring the situation and taking
some additional steps to lower the latency.
On Sep 28, 1:52 pm, App Engine Downtime Notify <
i...@google.com>
wrote: