docker – Switched from Heroku to Scaleway and server is much slower

I recently moved our backend services from Heroku (hobby-Tier (8€/month)) to Scaleway (DEV-L, 20€/month) as well as the database (similar prices).
We did this because of all the issues with the privacy shield and data security.

For this, we needed to containerize the application and I added some additional stuff (Grafana/Prometheus and Elastic Stack).

The problem is, that the server is much slower now. I mean A LOT. Before we had for very complex and large queries loading times of about 3-5 seconds, now sometimes like 20 to 25.

Do you have any idea what the problem could be? Or are the ‘dynos’ on Heroku just very optimized?
I tried shutting down all containers which aren’t needed (only Nginx, certbot, and the actual backend left) but that didn’t help. I also tried to upgrade the server or the database, which of course improved the performance, but only so little (max 20% for 150€) for so much money.

Additionally Grafana tells me that the server isn’t even used to capacity. Generally, the cpu usage is around 50% (most of the time) to max 125%, the backend itself never above 60 % (if there is a ‘complex’ request).