Azure File Share - Auto Scale Exclusion Hours

I was analyzing Azure File Share metrics in the Azure Portal and noticed that the “success server latency” metric was spiking daily. I realized that the Azure File Share has Azure Backup and snapshots enabled. I've observed that when the backup starts at 11:00 PM, the success server latency spikes significantly, triggering the auto scale to expand every day. Which in turn locks up the “once in 24 hours” and can never scale down because of the next deys backup spike.

Is it possible to add Auto Scale Exclusion Hours so that the auto scale isn't affected by something like a backup, especially if the spike happens after hours?

0

Comments (2 comments)

0
Avatar
Carl Long
We appreciate your feature request—community input is essential to our ongoing development.

Next steps:
     • We will review your suggestion and update its status during the evaluation process.
     • If further clarification is needed, we'll contact you via comments.

We also encourage others to contribute through feedback and voting.
0
Avatar
John Tokash

I really like the idea - I spot checked a couple of my larger customers, and couldn't find similar behavior occurring.   I haven't observed it, so the curious cat in me wants to know more. How utilized is your Azure Files Share?   I checked a small (200gb), medium (400gb) and one of our largest azure files shares for similar latency spikes when backups kick off and couldn't find evidence of the behavior.

I like the idea, to be able to set an exclusion time when auto-scale could ignore latency metrics, but on the other hand, I would want to know more about why the latency spikes at that time.  Its been a little while since I've read through the documentation on how the snapshots are captured, but I'm pretty sure Microsoft would contend you shouldn't see that from the backups.   

Since this isn't the place to collaborate on the topic I'll just stop there and say yes, this seems like it could be useful in some scenarios.

Please sign in to leave a comment.