when visualizing system.diskio.read.bytes I get only a growing chart of total bytes. This is not very handsome. I need a growth in read.bytes per rtt-interval (round trip time).
Is there any of the following solutions implemented I can use:
an option=delta in metricbeat which provides (additionally) the difference between actual read.byte and read.bytes from previous run (by keeping them in memory in the harvester) or
a pipeline function to access the previous value for same key-measurand to built a difference or
a scripted field function in Kibana to build a difference of 2 values in sequence or
a movavg(2) function in Kibana's visualization to calculate growth.
If not, are there any plans to implement any of them (option1+4 preferred). I imagine you also already noticed, that implementing the go diskio package is too less (otherwise you would have build a sample diskio bottleneck dashboard) and that building the difference is the key to success for diskio analysis.