This message was deleted.

Seems like one of the biggest cost savings can come from identifying and turning off virtual machines that do not need to be active.

for sure, but also making sure log analytics is configured to your needs and not monitor too many metrics will help (remember that Microsoft advice to drop two metrics to bring the costs back by half).

The process/latency/RTT metrics add up the costs big time, it's amazing how quick that balloons...

MS dropped the per process metrics by default https://techcommunity.microsoft.com/t5/azure-virtual-desktop/updated-guidance-on-azure-monitor-for-wvd/m-p/2236173 - for historical analysis / capacity planning log analytics seems pretty good but to get something useful with Monitor for proactive alerting is hmmm.... hard work and i really resent the pay per threshold model