Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Send to excel error

Hi Everyone,

I have an issue which needs your kind attention.

I have a straight table, has almost 555,000 records. This chart is eating up 334MB space. And when user tries to download it bysending it to excel or export it the application hangs. Also tried by reducing the number of records by making some selections. Still no help. Is there a way to allocate more memory for the chart.

Did anybody come accross such an issue. Please give me your solutions or valuable suggestions.

Thanks,

Bikash

4 Replies
khadeer
Specialist
Specialist

If ur user using excel-2003 version it has only limit of 65,536 rows by 256 columns, for 2007 1,048,576 rows by 16,384 columns. I think 334MB is huge size. So check it and try to reduce the number of records.

Regards,

Khadeer

jonathandienst
Partner - Champion III
Partner - Champion III

Hi

May I ask why you need to export 555,000 rows / 334MB to Excel. This is simply too much data for a straight export from the front end. I don't foresee many practical uses for Excel files that big (life is too short to wait while Excel loads/recalcs etc).

If you need a bulk transfer, use STORE in the load script into a csv file.

Otherwise, let the user reduce the number of records by making whatever selections and then allow the export, You can use conditional calculation to suppress the calculation of the table until the selections reduce the data to (say) 10,000 records or less. Also reduce the number of fields in the table, if necessary by splitting into 2 or more tables. If you are doing the export in a macro fired by a button, conditionally enable the button as well.

The expressions for the conditional calc/enable depand on your data model. For example, if you have a transaction ID, then you could use something like:

     =Count(Distinct TransID) < 10000

Regards

Jonathan

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

Hi Jonathan & Khadeer,

Thank you guys for your prompt response. I learnt this trivial things and did apply them as well.

Now where my question stand is even though I reduce the number of records to some 28000 rows still I am having the similar issue.

there are no synthetic keys, only two expressions in the chart that too a sum function, no set analysis nothing.

Regards,

Bikash

jonathandienst
Partner - Champion III
Partner - Champion III

Bikash

A couple of general things to look for:

  • Are you using a calculated dimension? These can be performance killers.
  • Are all the fields used in the dimensions and expressions in the chart in the same table? In large data sets, spanning across too many tables can result in out of memory errors. Even more so if the tables are nt asscoiated (island or orphan tables).
  • Are you using Sum(If...)) expressions? As long as the If condition can be evaluated outside, set expressions will perform far better.

If you post your dimensions/expressions (or your model if its not too large), we could be more specific.

Regards

Jonathan

Logic will get you from a to b. Imagination will take you everywhere. - A Einstein