The response time graphs changed substantially on a certain date.

While our monitoring system utilizes multiple nodes in disparate locations to confirm detected failures (primarily to avoid false alarms) the graphs are produced from a single node, and hence subject to this variability.

On the date in question, your host was re-allocated to a primary monitoring node in a different geographic location, hence the structural shift in the response time profile.

Such re-allocations are necessary from time to time to ensure the ongoing robustness of our monitoring system in all conditions.

Note that you can also specify your primary monitoring node for a given host.