Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi all,
I was wondering which approach out of Canonical date, Date Island and Link Table would result in faster reload of data and charts for our data. I have attached an image of how our data looks like. I appreciate any inputs on this.
Or could you please point me to an article that compares the pros and cons of these approaches?
Thanks!
for large data sets canonical would gie you best performance
check this blog series
https://optimalbi.com/crafting-qlik-sense-calendars-part-4-of-4-canonical-dates/
We do have a very large data set and we are currently implementing a link table style solution like in the article. I was just wondering if there is a more optimized solution
link table should work but canonical gives best performance as far as i know.
mark as answered if no more questions
Hi
Example. I had many dates in 4 joined (keep) tables. Three I used canonical dates for and table 4 a date Island approach.
I removed the date island approach. It actually worked well but
- It can't be used with a dimension.
- It's different. And I forgot after a time what I had done when asked to adjust the App (this was starting out)
So I tried a Canonical Date for this table 4 (I was using it for the other tables) but it got confusing as there wasn't a clear finer grain table (table 3 vs table 4). So table 2 was missing calls from table 1. Likewise table 3 from table 2. This was overcome by making table 3 the finer grain table and adding the missing call numbers to table 2 and 3. But table 4 was missing calls from table 3 and table 3 likewise from table 4.
So I simple concatenated table 4 to table 3
My view is either use canonical dates alone or canonical dates after concatenating tables.