Skip to main content
Announcements
Qlik Connect 2024! Seize endless possibilities! LEARN MORE
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

object out of memory

Hi,

Due to increasing data some of my charts are showing message like''Out of object memory''. what is the possible reason for this. can any one advice what to do in such particular condition.

Regards,

Piyush

6 Replies
Anonymous
Not applicable
Author

This means that you don't have enough RAM on your machine to run the object calculation. Reduce that data and then check. Also, try to optimize the calculation, use less if conditions, use set analysis.

Regards,

Aadil

Siva_Sankar
Master II
Master II

Can be fixed by having proper data model.

Check in the sheet properties, object tab, which object is consuming more memory.

Debug the object why it is consuming more memory,check the expressions if it is complex, try using it in script side.

Lock selection if the dashboard uses year, by selecting always one selected value in the listbox.

Try implementing optimization techniques...refer

http://community.qlik.com/message/438756#438756

regards

siva

NareshGuntur
Partner - Specialist
Partner - Specialist

The CPU will do all the calculations(expressions, set analysis etc).

If you can't change the expressions/data model, you can use the calculation condition so that the current chart(out of object memory) will be not get calculated until the condition is fulfilled. In this way, that particular chart will not get calculated along with the other objects so that all you CPU will calculate your current object.

Cheers,

Naresh

jonathandienst
Partner - Champion III
Partner - Champion III

Hi

This is often caused by expressions that contain fields from different tables in your data model. This causes QV to perform a join on these tables in memory to perform the calculations. If both of these tables are large, this join can consume large amounts of RAM.

The solution comes down to one or more of the following:

  • modify your model. Possible denormalise the problem field(s)
  • modify the expression (look for inefficient/expensive constructs such as distincts and sum(if())
  • use a calculation condition so that the filtered table size (ie after making selections) is acceptable
  • pre-calculate or pre-aggregate the expressions (or parts of the expression) in the load script
  • increase your RAM (although you're probably going to hit the limit again very quickly)

HTH

Jonathan

Logic will get you from a to b. Imagination will take you everywhere. - A Einstein
Not applicable
Author

Jonathan,

How would you check the filtered table size?  I've seen checks for whether getselected is within a certain boundary, but in truth just requiring selections that reduce the total table size would be ideal.  Just not sure hoe to check for that.

greg-anderson
Luminary Alumni
Luminary Alumni

While it is possible that you just have too much data, it is more likely that you have one or more of the following:

1) a synthetic key in your AQL data model

2) two unrelated data elements with the same name, causing false relationships

3) complex calculations in your front-end objects.  Put them in the script as much as possible.