I am using Grafana to display data from AWS Timestream, it seems to be very slow to display the data, I have tried changing some datasource configuration but the situation remains the same, it is really very slow, do you have any ideas to improve performance ?
The query extracted from the panel and run on the timestream query editor takes 2 sec (see image)
Reloading graphana after a few minutes, it started the slow loading again, and opening Chrome’s DevTool you can see how the query calls repeatedly fail until an answer comes after several minutes (as you can see in the second pic)
i’ve used try_cast because not all values are integer inside the Timestream table (and it’s useful to display those type of data in Timeseries visualization) and no, i don’t have indexes for those columns (dimensions). At least with also some simple queries the situation remains the same
yeah that was just some side points but worth looking into. can you try that same query without the use of try_cast
you should really not be plotting any values that are not int though right? also it raises a lot of ??s as to why that one column has 2 types of data types. does not pass the sniff test
Without the use of try_cast the situation remains the same, it seems that the authorization provider via SDK takes a long time before it has a token to be able to query AWS services and every time it expires it takes the same amount of time to request it
