Slow Qlik document

[pauldamen]

All,

I have a Qlik document which loads from 3 QVD's (around 500.000 lines in total).

And I have 5 tabs with some objects, my problem is, it is extremely slow. Every time I make a selection it eats up all my CPU and takes forever to change.

What did I do wrong that makes the document that slow?

Regards,

Paul

[datanibbler]

Hi Paul,

I'm downloading your document, but our internet_connection is a bit sluggish at times, it might take a while 😉

Just as a first hint - when the app loads acceptably fast, but every selection causes it to work for a while, it means that your object/ chart is too complex. That might have to do with calculated dimensions or a nr. of other issues.

=> If you have calculations in your chart, try to move as many of them as possible from the GUI into the script (there they are executed only once, not with every new selection).

Sorry I cannot give you a more precise hint right now, I'd have to look at your .qvw.

HTH

Best regards,

DataNibbler

[rustyfishbones]

Hi Paul,

It's the same when I try to Open the file, really slow.

Can you reduce the Data by only loading FIRST 1000 rows for each QVD and repost the reduced App.

Or can you post an image of the Table Viewer

Regards

Alan

[pauldamen]

See first 1000 rows load for each QVD attached

[ashfaq_haseeb]

Hi,

I have seen your both files.

Its because of heavy calculation you used in fronted.

I suggest to take all calculation (As much as you can) to your script.

Regards

ASHFAQ

[pauldamen]

My calculations contain a lot of sums, but that doesn't work in the script?

How can I solve that?

[tresesco]

Your front end expressions are too complex. Try to optimize it. One hint:

Instead of : =Sum(A)-Sum(B)-Sum(C)- ....  try

                   =Sum(A-B-C)

[rustyfishbones]

Hi Paul,

You Customer as Dimension in some of the Charts and there are 346,080 this could be causing the speed issues.

Also the SubsetRatio of Main Dim Table is 3 % this won't help you either, you have rows in your Dimension Table than you do in your fact table, so you have lots of Customers that don't have facts you need to remove these in your script

Can you share your qvd's and we can take a look at creating a better Data Model perhaps

[rustyfishbones]

Hi Paul,

Change your script in your main tab to below and reload the document.

Let me know if the speed increases

SET ThousandSep=',';

SET DecimalSep='.';

SET MoneyThousandSep='.';

SET MoneyDecimalSep=',';

SET MoneyFormat='€ #.##0;€ -#.##0';

SET TimeFormat='h:mm:ss';

SET DateFormat='D-M-YYYY';

SET TimestampFormat='D-M-YYYY h:mm:ss[.fff]';

SET MonthNames='Jan;Feb;Mar;Apr;May;Jun;Jul;Aug;Sep;Oct;Nov;Dec';

SET DayNames='ma;di;wo;do;vr;za;zo';

Main_Exp:

LOAD Date,

     [Rateplan Sim Only Grouping],

     CUSTCODE,

     Revenue,

     Discount,

     SaldoTotal,

     GrossAds,

     Renewals,

     CreditAmount,

     IAS,

     VOLUME_COGS,

     [IAS Name],

     Rate,

     COGS,

     Code,

     RateMI,

     #Customers,

     UnbilledAmount

FROM

Main_Exp.qvd

(qvd);

Main_Dim:

LOAD CUSTCODE,

     MainAccount,

     Customer,

     HOLDING_KEY_NEW,

     Channels,

     [Ds Ids],

     CUSTOMER_ID,

     RatePlanCode,

     RatePlanName,

     Segment

FROM

Main_Dim.qvd

(qvd)

WHERE EXISTS (CUSTCODE);

Temp_1:

LOAD Type,

     Date,

     ExtraCustomerAmount

FROM

Temp.qvd

(qvd);

[pauldamen]

No it doesn't help