Is your feature request related to a problem? Please describe.
I've noticed that our stories and examples are using fake datasets with low to minimal meaning.

Describe the solution you'd like
I'd like to change that in favor of a meaningful dataset extracted from a Kibana instance.
I've done some experiment with arquero and I feel it is very powerful and simple to handle data query/aggregation and can replace most of our math/agg code and can be systematically used on every example to query the and get the right amount of data we need.

Describe alternatives you've considered
Build and use a set of function to aggregate/filter a dataset we own. This just requires more effort than reusing an existing library.

Additional context
With real case datasets we can find/handle more edge cases than we have done until now.

Kibana Cross Issues
n/a

Checklist

• this request is checked against already exist requests