Forum Discussion
how to improve cumulative performance
- 3 years ago
finally i found the solution
oke, the issue is "slow performance while using cumulative"
root couse -> my chart using "dimdate" instead of "logdate"
solution -> for cumulative dont using relational table, use column that u call in measure
i change from dimDate to logdate and viola from 1 minutes to 15seconds
Hi nofan_munawar,
How many records stored in data table that you calculated? Can you please share some mode detail about your scenario?
How to Get Your Question Answered Quickly
It obviously will cause the performance issue when you use cumulative expression to calculate through huge amount of table records. You can also try to use the following formula if helps:
cumCustomerByBook =
VAR maxDate =
CALCULATE (
MAX ( SellingUnit[logDate] ),
ALLSELECTED ( SellingUnit ),
VALUES ( ClusterMapping[clusterCode_unit] )
)
RETURN
CALCULATE (
SUM ( SellingUnit[counting_unit] ),
FILTER (
ALLSELECTED ( SellingUnit ),
SellingUnit[logDate] <= maxDate
&& SellingUnit[flag] = "Customer"
&& SellingUnit[isLaunching] = 1
),
VALUES ( ClusterMapping[clusterCode_unit] )
)
Regards,
Xiaoxin Sheng
- nofan_munawar3 years agoFrequent Visitor
finally i found the solution
oke, the issue is "slow performance while using cumulative"
root couse -> my chart using "dimdate" instead of "logdate"
solution -> for cumulative dont using relational table, use column that u call in measure
i change from dimDate to logdate and viola from 1 minutes to 15seconds