Hi,

I would be interested in transaction\sec and batches\sec.
Along with an idea of what is actually running at that point to cause the spike.

Is it more transactions or bad code or process.
Profiler might be able to point you in the direction of what code is running.
Or you might want to use SP_WHOISACTIVE and turn on logging.

It would seem to me something runs or was running in that 10-12 min period which had a drain on resources.

SC