> 1. Can we activate the timer only when requests are actually pending? > Imagine a host with 1000 guests, even a 1 second timer becomes > wasteful.
I will try to do this. > 2. You don't vary the wait time, does this mean a throttled request must > wait for max 1 second? If yes, then it introduces a big variance on > request latency. If iops or bps request where done on a negligible time (very fast storage backend) yes. We could make the timer frequency higher though to mitigate this. Best regards Benoît