Report Performance

[former_member486909]

Hi All,

Can creating more and more aliases of the same data-source deteoriate report performance? My report has a single data source but 4 aliases. Does this affect performance?

What are the other factors to be kept in mind to improve report performance?

Thanks & Regards

Swasti

[Former Member]

Hi Swasti,

The Answer to the Question is No it is actually an advantage to have such Aliases(this also applies to the .

The way Design Studio works is that, once a connection is established to the Query, which is the source of the Datasource then it connects itself on the samething and it does not take as much as it takes the first one.

For e.g. DS_1(the first alias) takes 320 ms.

              DS_2 takes only 20 to 80 ms. and so on.

Regards,

Fazith Ali Z

[Karol-K]

Hi Fazith Ali Z,

can you please elaborate in detail on the performance impact of data source aliases?

why are you saying "no" if already in your scenario you have additional 25% degression on performance on second alias?

I would say - number of aliases has a big impact on performance and the scenarios should be created with the goal ro restrict this number.

Regards, Karol

[Former Member]

Hi Karol,

There were conditions where in our Dashboard where we had to use multiple Data Aliases of the Same Query. One Example is this,

We had a View which had 6 different charts. in the Sixth chart there is a Toggle between a Pie which showed overall data and then by trend,

couple of other charts show only trend coming from the same Query.

In this situation you could only use two data source to solve this problem.

There is also one issue, at times when you use Swap/add/remove Dimensions with swap chart type, the Application throws up an error.

So in these situation, it was safer to use multiple Data Aliases.

My Answer for the Question, is referred to the above scenario!

Regards,

Fazith Ali Z

[Karol-K]

I understand that there are scenarios you need more data sources and more aliases.

Only in terms of performance impact, the correct answer is YES, this has impact on performance. sometimes bigger, sometimes smaller - but always

[Former Member]

HI Karol,

My Answer was only based on that Scenario, and in almost all of our client Dashboards, we have this scenario. The no. of Datasource DO have an impact on the File.

But my Question is, if it is just 20 to 80 ms. How does that matter in the overview, when the entire thing is just gonna get delayed by 1/2 a second.

I am asking this in the View of a Developer/User.

Regards,

Fazith Ali Z.

[Karol-K]

the time you observ is very dependent on the scenario, you simply cannot predict how much the effect of new data source will be. in small data sources, quick BW backends it will be very small. but in bigger data sources it can increase quickly.

bigger degregation will be visible if you have variables, and you need to change one - then it will affect many data sources and all must be restarted again.

I agree, in small resultsets cases the effect is also small, so it is always a question of the scenario - you need simply to check if the complete scenario is still acceptable.

Karol

[Former Member]

Hi Karol,

Thank you for your Advice, I will make sure I have them in mind next time I'm Developing the Dashboard.

Regards,

Fazith Ali Z.

[TammyPowlas]

Please search before you post again in the future, per the SCN rules of enagement. 

There have been documents on and blogs on this very topic.  As a start see by