Question on data source and data source view of MOLAP cube

Hi, all experts here,

Thank you very much for your kind attention.

I am having a question on the data source and data source view of MOLAP Cube. As we know MOLAP Cube stores all data in MOLAP format, it does not have any thing to do with its source data base. As a result, I am wondering if we could actually remove the data source and data source view of the MOLAP cube? (in my case, no update is necessary to be informed from the underlying data source to the MOLAP cube)

Hope it is clear for your help.

I am looking forward to hearing from you for your advices and help.

With best regards,

Yours sincerely,

Hello again Helen.

You will always need a data source and a data source view for the cube.

What you can do, in order to create an off-line cube is to use local cubes, if this is the scenario you are thinking about?

With local cubes you create a local cube file that you can put on a laptop and analyze data without being connected to SSAS2005.

Both ProClarity and Excel2007 have this option.

HTH

Thomas Ivarsson

|||

Hi, Thomas,

Thank you for your very kind advices.

But I found that change of the data source view does not really reflect on the data source view of the cube though? I mean after I created the cube, and I edited the data source view, but when I clicked on the cube, its data source view remained as the one before I changed to the new data source view?

Also, I am a bit confused, as all the data for the cube has already stored in SSAS2005 Data base (MOLAP), so why we still need to care about the data source and data source view if we dont need to have any update from the underlying data source?

Thank you for your kind attention and I am looking forward to hearing from you shortly for your further advices.

With best regards,

Yours sincerely,

|||

Hi, Helen.

I am not sure about the scenario here? Do you mean that you will create the cube once and never process it again?

If new dimension records and new fact records arrives from the data source you will always need to process the cube. If you need to process you will need to have the data source view.

During a days work new transactions will be registered in the source system. You will have to do a new process of the cube in order to get them from the source system into the cube.

Are you making cubes for laptop users that they will use off-line? Or are you building cubes using real time OLAP?

Regards

Thomas Ivarsson

|||

Hi, Thomas,

Thank you for your reply. In my case, well, I do deploy the whole project containing the cube after I changed the data source and data source view, but the cube remains the same as before I did any change. Therefore, I thought data source and data source view does not have anything to do with the MOLAP cube after its first deployment to the analysis services server.

Does it work in this way? Or I miss out anything of it? Thank you and your advices are most welcome.

With best regards,

Yours sincerely,

|||

Hello Helen. If you make a change like adding an attribute in the data source (source table/view) or the data source view(=named calculation) this change will not be a part of the cube until you add the attribute in the dimension editor.

In order to make a change in a cube you must mapp new columns to the measure group or to the dimension in the cube/dimension editor.

The data source view do not automatically update the cube or the dimensions. If you do a refresh in the data source view(right click) all changes in the source database will exported to the data source view.

HTH

Thomas Ivarsson

|||

Hi, Thomas,

Thanks a lot for your kind advices.

It cleared my confusion over the above question.

Thanks again.

With best regards,

Yours sincerely,