How to best track changes to large numbers?


(Scott Chapman) #1

I am currently using Kibana to kee track of kafka offsets. Offsets come in
a triple of numbers; size (number of messages posted to topic), offset
(where in the list the consumer is) and lag (difference between the two).

One thing I would like to be able to visualize is the increase in the size
over time. The problem is that it is a pretty large number, and the
increase is kinda small compared to that over any period of time.

My only options are Sum, Max, Min, Average, STD, and such. I don't see a
good way to visualize the rate of increase, or change in the value over
time.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/1099d2b2-502e-4f9b-8576-248195e0ec41%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


(charlesmims) #2

Did you ever find a solution to this problem?
I've been working on a similar problem myself, wanting to show packets per second across an interface as reported by collectd; the metric as sent to logstash is a cumulative total since the machine was started. The approach I've been trying to get working is using ruby filters to subtract the previous count from the current count, set a global variable to the current count, then set the the current count to the difference of the previous and the current counts. I've had mixed results so far, but I'm blaming my lack of ruby knowledge, or possibly receiving updates out of order at this point.


(system) #3