Do proxy data-sources (that is, queries being served by Grafana acting as a proxy) allow any kind of caching?
I’m interested in having caching for how the data sources are queried, since my company has many different people monitoring the exact same things, thus repeating the same queries to the databases. Worst case scenario, sometimes something bad happens and «everyone» jumps in into the dashboards overwhelming the database with queries which are merely repeats of one another.
No, grafana does not do any caching currently
Would that qualify as a reasonable feature request? Shall I open it as a feature request?
Or should that be regarded as something out of Grafana’s scope?
think there already is a feature request
Hello, has the situation changed pertaining caching?
In particular, does Grafana support any type of caching to avoid repeating heavy MySQL queries, or would it be preferable to have a caching mechanism feeding Grafana for frequent (computationally-heavy) queries?
Is the only option something such as views of desired statistics in MySQL or caching through Apache?
guess still no update on this?
In my opinion this “feature” does not belong in Grafana, but instead in the
back-end data store.
The reason is simple - Grafana cannot tell whether a query which is run more
than once is going to return the same result, or different results, on each
run. The background data may have changed between queries, and in that
instance caching and re-using the previous result would be inaccurate.
The back-end data store, on the other hand, has the ability to know whether
data has been modified between queries, and can thereby re-run the query with
the new data, or simply return the previous result if the data hasn’t changed.
Regards,
Antony.