Too many objects on one sheet causing performance problems?
I'm creating a dashboard that has essentially multiple "pages" in one tab. There are a set of buttons that set a variable to a different value. There are multiple sets of charts that use the show condition on the variable to show/hide. This way I can have a dynamic dashboard with lots of content without having to flip through different tabs.
I'm seeing some performance problems though. It's just really slow to interact with. I'm wondering if this is due to having too many objects on one sheet. Is there a difference in terms of performance between having objects on one tab instead of multiple tabs? Does QlikView load only the objects into memory that are on a single tab? If objects are hidden, do they still take up memory?
As well as setting an expression for Show Conditional, also put the same condition in the Calculation Condition of the General tab. This will stop them being calculated when they are hidden.
I would recommend hiding the tabrow and instead of conditionally showing/hiding, active different tabs with the required charts (keep the tab design the same for the look and feel with just charts different). This means you can have less objects per tab, which should improve performance, plus you can remove the conditional shows overhead too.
You can use variables and actions to show and hide tabs giving the appearance to users that there is just one tab, whist they are actually switching between a number of different tabs.
Use the variable to control the Sheet Properties, General, Show Sheet, Conditional expression.
You can have several sheets that have different Sheet IDs but the same name in the Sheet Properties so will appear as different versions of the same sheet.
This approach could reduce the overhead of calculating the objects.
That would save me a lot of time. Would adding the calculation condition free up memory the same way splitting tabs would? I've already created dozens of charts so splitting everything into tabs will be a ton of work.
