Grafana v9 slow after adding alerts

I’am using Grafana on Kubernetes with the default configuration (except for the root_url, domain and cache_mode parameters). This Grafana is getting data from 2 Prometheus deployed on the same K8s cluster.
Lately, I upgraded my Grafana to v 9. When I’am not using Alerts, my Grafana is running perfectly fine. But as soon as I add some alerts, it is getting really slow to do anything: opening dashboards, getting rules, getting data, … I’ve checked and it does not come from a resources issue.
Does someone know what is going on and how to fix it ?