SSAS Cube Partition
Let’s start the cube partition now.
What is cube partition?
It is the process of breaking the bulk measure groups into multiple chucks of data. For instance, if you have measure group with sales data of various dates then you can keep the measure group data partition with data range (Jan-Mar, Apr-Jun…)
How to create a cube partition?
Open Analysis Service solution then open the cube design and go to partition tab as shown in the below picture.
Before creating a new partition, we have to change the existing source to Query binding type. It helps us to create multiple partitions with restricted rows.
After completing the above step, you can create new partition by clicking create new partition button or link as shown in the below picture
Now, Partition wizard will show up.
This is just a welcome screen. You can read the text given and Click next.
In this screen, you have to select the fact table which you want to create cube partition.
In this screen, you have to select a check box “Specify a query to restrict rows” and add where condition to the given query in the text box with the range field. For instance, if you are using date rage partition you can use the date key field in where clause. Refer sample query in the below diagram. Click next to move forward.
In this screen, you can specify the processing option and the storage location of this partition. For now, you can use the default option and click next.
This is the final step for creating a partition. In this screen you can name the partition of the cube and specify aggregation option. For now, you can select “Design aggregations later” option and use default cube name. Click finish button to complete the partition creation process.
Finally, you will see the cube partition definitions in the partition list as given in the below diagram.
I have used adventure work database and the cube design in this article. Thanks to Microsoft.