summarize performance

[ericjo]

hello, 
I made a measurement using the summarize function as below, but the speed is slow, so please help me if there is any way to improve it.

a : Cust_ID Summarize with one exter_id (Order_ID) based on the customer type "Rewards" based on Cust_ID
b : Check the quantity purchased by product using cust_ID of 'a'

# Purchase_1_A = 
var a =
FILTER(
SUMMARIZECOLUMNS('anl_dm_loyalty_sales_point_mart'[saguid])
                            ,CALCULATE(DISTINCTCOUNT('anl_dm_loyalty_sales_point_mart'[exter_id])
                                        ,'anl_dm_loyalty_sales_point_mart'[Cust_Type] = "Rewards"
                                        ,ALL('std_product_loyalty'[product_name])
                                        )=1)
							
var b = 
SUMMARIZE(CALCULATETABLE('anl_dm_loyalty_sales_point_mart'
                                        ,'anl_dm_loyalty_sales_point_mart'[Cust_Type] = "Rewards"
                                        ,'anl_dm_loyalty_sales_point_mart'[saguid] IN a
                                        )
                            ,[saguid]
                            ,[obj_id]
                            ,"Qty",CALCULATE(SUM('anl_dm_loyalty_sales_point_mart'[quantity])
                                                ,FILTER('std_product_loyalty','std_product_loyalty'[product_name] <> BLANK()))
)
                                                
Return
SUMX(b,[Qty])

[tamerj1 · Accepted solution]

Hi @ericjo 
Any restrictions of using GENERATE function?

[ericjo]

I will write Dax using GENERATE function.

The aggregate should be different depending on the date Between selection in the report, so we made Summarize from the Measure value, but is this the part that is considered?