For following use case, I’m not sure about best approach:
I load data from a REST-Api per minute and save them to influxdb. It’s about 99 different datapoints per minute (mixture of string, integer, float).
Example:
dp01 = ‘foo’
dp02 = 123
dp03 = 42,1
…
dp99 = ‘state’
Next day I receive a CSV-File with additional datapoints per minute for the last day, which are not in the REST-Api.
Example:
dp100 = ‘foo1’
dp101 = ‘foo2’
Should I simply write the CSV-datapoints dp100,dp101 next day into influxdb, or should I already add them with empty values durung the REST-Api processing.
Example
dp01 = ‘foo’
dp02 = 123
dp03 = 42,1
…
dp99 = ‘state’
dp100 = ‘’
dp101 = ‘’
Not sure about indexing inside influxdb or other consequences on performance.
Thanks for advice.