As described here, I want to implement the "rescaling" method in Kibana.
I have a numeric field, which is always >= 0 .
I want it to scale to the range of [0, 1].
It seems:
Timelion doesn't support extracting the Maximum bucket / Overall max. Similar to this
Kibana Time Series does support Overall Max, but when using a calculation aggregation, I can't use the Overall max aggregation.
Is that correct? If yes, are there any suggested workarounds?
Thanks
Also, just to point out, the Wikipedia article you linked to seems to suggest calculating the scale in a pre-processing step, so it probably wouldn't hurt to originally index the data with a scaled field.
Actually, I intend to use the Overall Max metric in the Calculation and not the other way around.
Calculating it pre-index time won't work as I need.
That is because the calculation depends on the Overall Max and Overall Min, which depend on the time range you are using.
If I want to be able to easily change the time range I need it calculated every time.
@Eran_H Unfortunately Time Series Visual Builder is limited to the functionality that Elasticsearch provides for the bucket_script (which is what powers the calculation agg in TSVB). We are unable to use sibling aggregations (which is what the overall max aggregation is) in buckets scripts because how of the calculations are processed during the reduce phase.
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.