Performance issue with 20k dimension rows

the problem is as follow : 

 

i have a dimension table (20000 rows ) dimaccounts   and a fact table (  2 million rows )  fact_sales

 

im building a pareto chart : 

 my DAX measure : 

 

DEFINE

MEASURE 'fact'[cumu] = 

var _current_sales =[sales]
var ds = 
	filter(
		ADDCOLUMNS(
			ALLSELECTED('dim'[Value]) , 
			"@sales" , sum('fact'[Value])
		) ,
		[@sales] >= _current_sales
	)

var res =  sumx (ds  ,[@sales] ) 


return res  

 

 

[sales] measure is a simple sum (fact_sales[amount]) 

 

my approaches : 

1.  visual calculation --  performance issue. 

2.  rankx in code --  performance issue 

3.  window function --  huge performance issue. 

 

explanation of the measure : 

it is calculating the cumulative sum of sales, ordered by accountid .

so if i have per example : 

accountid         sales

1                           20

2                           40

3                            10

 

the visual should show  the cumulative as below : 

accountid            sales    cumulative

2                            40            40

1                             20           60

3                             10            70

 

 

the problem is that the dax measure is taking like 50 seconds to finish : 

 

 

 

and im stuck into finding another way to caluclate the value that can improve performance. 

 

Any help on this would be appreciated. 

 

Note : even with a fact table of 20000 rows will take a long time to finish exeucting the dax code.