If I understand you correctly, then you can anything not to change in the accumulation of your data in one large table. Further when constructing the necessary charts, you can apply selections, bookmarks and use in diagrams of Set Analisys expressions to obtain the desired result.
I should clarify. So the user interface should be simple - meaning the data needs to be easy to consume (aka: should not need to devolve to set analysis to do multivariate drag and drop). Its plausible to change how data is injected, but there are bigger issues (10,000 columns...), so not an immediate solution there.
Given this, then the question is if at ingestion Sense can break out the rows into columns in the QVD, and hence resolve the core issue.
Anyone have a thought? Kind of a pivot import (rows to columns)? So from the above to the below:
[Type] [parentID] [quantity of electricity | ampere second] [electric flux (flux of displacement) | ampere second] [angular repetency | radian per metre] measurement x .0000999988888 measurement z 1.232324 measurement y 121.56434566
have a look in the help Qlick Sense Help section for Generic load :
So have a look here also
This document helps in understanding Generic Load concept which pivots/transpose given data.
hope it helps