Optimizing loading times in dashboards with Prometheus data source

  • What Grafana version and what operating system are you using?
    Grafana v9.4.7 (but we can upgrade anytime), container version on Debian host
  • What are you trying to achieve?
    I am trying to optimize the loading times in my Grafana dashboards by reducing the redundant calculations of the metric avg_over_time(probe_success[$__interval]).
  • How are you trying to achieve it?
    I am looking for a method to calculate the metric value only once and then use it across multiple panels in the dashboard. I am considering the use of variables or other techniques to achieve this optimization.
  • What happened?
    Currently, each panel in the dashboard independently executes the avg_over_time(probe_success[$__interval]) metric, resulting in redundant calculations and significantly increasing the loading times of the dashboard.
  • What did you expect to happen?
    I expected to find a way to optimize the loading times by performing the calculation of the metric only once and then reusing the value across all panels in the dashboard. This would help reduce the overall loading time and improve the performance of the dashboard.