Are the kibana dashboards used for the recent "Elastic for Operational Analytics" webinar available somewhere?
I am in particular interested by the dashboard shown at 12min and the distribution per country.
I have a relatively straightforward index that stores requests to a service and there is a requestCountry parameter there with the country code. How can I create a visualization that is similar to the one shown in the webinar?
We do not redistribute the dashboards themselves, since they are pretty specific to the example data used. But I can point you towards some blog posts that describe several use-cases for geographic data and coordinate maps:
Last, but not least, version 5.5.0 of Kibana added another map type, the region map, which is well suited for abstract or custom geographic visualizations. It sounds like your requestCountry field could be mapped quite easily to such a region map.
Thanks for the feedback. Actually I was mostly after the "Unique visitors per country" graph. Could someone describe what graph type this is and how it was build?
I don't have any geoIP information (besides the country code at the moment) and region map may be a nice alternative. We're in the elastic cluster and we're running 5.5. I had a quick look and couldn't figure out how to map that to the country code. I am happy to dig more but that link doesn't seem to provide me enough information to get started.
Sorry for misunderstanding you there. The graph looks like it was created using the Time Series Visual Builder, a specific visualization type. In addition to the documentation we recently published a few tutorial videos on using the Visual Builder:
With the Time Series Visual Builder, a graph like the one you mentioned can be easily created by combining the "Cardinality" aggregation on the field identifying each visitor (e.g. sessionid) and a "Terms" aggregation on the country code as "Group By" criterion:
That's awesome, I managed to create a graph on my data.
However, I do not understand the semantic of the numbers on the right. What are they supposed to mean? With a large period, it usually shows nothing and when I drill down I have some data but I cannot make any sense of it.
In my case, I'd live this to be the percentage of the total count (in your example above, 3% of the data comes from the USA).
Can you clarify what those numbers mean and if there is a way to show the count percentage instead?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.