cancel
Showing results for 
Search instead for 
Did you mean: 

Infocube and Multiprovider

Former Member
0 Kudos

Hi,

Please clear my doubt.

I have an infocube IC1 with data. I have a multiprovider which contains the same infocube IC1. If we query on both of them seperately, which one will execute faster and why?

Thanks in advance,

Sekhar

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hi, Performance point of view both will take same time to execute query ...beacause the multiprovider is having a single cube and even there no pint of using this if there is single object. if there are more than one target used in the multiprovide then the query run time will be slower that depends on filters/conditions/no. of records.

hope it helps.

thanks,

Vachan

Answers (4)

Answers (4)

Former Member
0 Kudos

The actual DB query on the single cube should be identical to the DB query that gets spawned from the one cube MultiProvider.

The main reason you would want to defeine a MultiProvider on top of a single cube is to give you more flexibility in the future.

Let's say you have IC1 and 20 queries, but now you have a new project and need to create a new cube, IC2, that is similar to the current cube but will have additional chars, dims, etc. This new cube should eventually replace the old cube. If you created your queries directly on the base cube, now you would have to copy the queries, and if you were using Bex Browser, you would have to notify all your users to stop using the old queries and begin using the new ones - same thing might happen with URLs for web queries that people might have bookmarked. This can start to be a big headache in a large organization.

If you had first defined a MultiProvider and created your queries against that, all you would have to do is modify the Multi-Provider to reference data in the new IC2 cube instead of the IC1 cube. The users would never have to make changes - they would never know anything has changed.

Always building your queries on top of a MultiProvider offers to potential to insulate the users form changes down the road. This is often considered a "best practice" in presentations I have seen.

Former Member
0 Kudos

Sekar,

Two things. If you have data only in IC1 , you can apply filter in the query with this cube. This way performance wise there wont be much difference.

Ravi Thothadri

Former Member
0 Kudos

I desagree with Suresh Reddy. There several points to use multiproviders on single infocube.

I think there is no performance diference bettwen both.

Former Member
0 Kudos

Hi Sekar,

There is no point of having Multiprovider for a single Cube. If you need a report which requires information from multiple infocubes then you need to tnion them in Multiprovider.

Any way for your question, The query executes faster on Infocube rather Multicube.

Thanks. Assign points if helpful