In the demo to show the performance difference between Tabular and Multidimensional, it seems to be returning all the fact records to all the dimensions in the Multidimensional example: all the regional subtotals have the same value. Could this be the case? PS: Thank you for this video. It was very useful as a first approach to this discussion.
That Multi Dim example is probably a bit bust for a head to head: see the values are the same across States? Seems you have missing links in your model with the measures, so I imaging "cross joins" happing behind the scenes (I know that's not the right terminology, but badly modelled data results in bad performance too!)
Excellent teaching SIR
Great one Alan that really helpded. Would you be able to share those additional articles for referring in the last slide of your PPT
Really amazing presentation it's so helpful thank you
In the demo to show the performance difference between Tabular and Multidimensional, it seems to be returning all the fact records to all the dimensions in the Multidimensional example: all the regional subtotals have the same value. Could this be the case?
PS: Thank you for this video. It was very useful as a first approach to this discussion.
That Multi Dim example is probably a bit bust for a head to head: see the values are the same across States? Seems you have missing links in your model with the measures, so I imaging "cross joins" happing behind the scenes (I know that's not the right terminology, but badly modelled data results in bad performance too!)
Microsoft is retiring Multi dimensional cube, if you just started you should learn Tabular only
So basically which one is better to use in the next 10 years, multidimensional or tabular?
@@norpriest521 tabular , Microsoft is retiring multi dimensional model