I am collecting switch port counters (octets) and graphing them as the NonNegative derivative(). I am trying to figure out a way to either:
- suppress the ports with the least amount of traffic (almost 0)
- only graph the port data with the most traffic, for various definitions of ‘most traffic’. Highest peak rate would probably be best.
I looked at top(), and highestmax(), but they seem to only return the record, not the entire table. I would like to maintain the accurate graph of the traffic
Ugh…um finding local min and max is non trivial in Flux and I wouldn’t recommend it honestly.
I tried finding the post where I achieved this but yah I don’t recommend going down that rabbit hole.
I honestly would suggest using the python client libraries and pandas to do this type of data transformation work.
I’m sorry I don’t have a better answer.
This was both a very helpful and slightly frustrating response. I think you’re probably right about python+pandas. However, I feel like a purpose built language like flux should have a builtin function to abstract the complexity.
All that being said, it seems that fluxlang might not be relevant much longer based on the direction influx 3 has taken. I guess the follow up question would be do you think it’s likely/possible that influx 3 will support any of the bespoke python+pandas or will I be rewriting all my queries yet again?