Hey! 
First of all thanks you so much for the effort and the brilliant work that you do for us. We sent two emails to Fly.io support on last Friday and Yesterday without any response.
Recently, we have deployed our application and we noticed response times experience spikes of 3-4 seconds. This issue occurs randomly across different instances in various regions (One particularly strange case we’ve noticed is that within the same region, one instance consistently maintains very low latency while another often experiences spikes of 3-4 seconds.). These spikes are negatively impacting our user experience and overall service performance.
We have attempted to address this issue by repeatedly adjusting the concurrency settings, but to no avail. Despite our efforts, the spikes persist. Here are some additional details that might be useful:
- The app is distributed in the following regions: ams, sea,mia,nrt.
- We are using nodejs for our application.
- We have monitored the logs and performance metrics but have not been able to identify a clear pattern or specific cause for these spikes.
could someone help us?
Thank you for your attention and support.
Best regards,
Luigi