We've captured the moments from FabCon & SQLCon that everyone is talking about, and we are bringing them to the community, live and on-demand. Starts on April 14th. Register now
Hi !
How you would optimise measure below? I tried to replace FILTER on KEEPFILTERS but performance stays the same.
Final_measure =
SWITCH (
SELECTEDVALUE ( 'Table1'[Column1] ),
1,
SWITCH (
SELECTEDVALUE ( 'Table3'[Column0] ),
"Reve", CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text0" ) ),
"PPO", CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text1" ) ),
"KO",
CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text2" ) )
+ CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text3" ) )
+ CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text4" ) )
+ CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text5" ) )
+ CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text6" ) ),
"PKO", CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text7" ) ),
"EBIT", CALCULATE ( [Measure0], FILTER ( 'Table2', 'Table2'[Column0] = "Text8" ) )
),
0,
SWITCH (
SELECTEDVALUE ( 'Table3'[Column0] ),
"Reve", CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text0" ) ),
"PPO", CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text1" ) ),
"KO",
CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text2" ) )
+ CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text3" ) )
+ CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text4" ) )
+ CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text5" ) )
+ CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text6" ) ),
"PKO", CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text7" ) ),
"EBIT", CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text8" ) )
),
SWITCH (
SELECTEDVALUE ( 'Table3'[Column0] ),
"Reve", CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text0" ) ),
"PPO", CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text1" ) ),
"KO",
CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text2" ) )
+ CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text3" ) )
+ CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text4" ) )
+ CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text5" ) )
+ CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text6" ) ),
"PKO", CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text7" ) ),
"EBIT", CALCULATE ( [Measure2], FILTER ( 'Table2', 'Table2'[Column0] = "Text8" ) )
)
)
In my opinion, Final_measure =
SWITCH (
SELECTEDVALUE ( 'Table1'[Column1] ),
1,
xxxx,
0,
xxxx,
xxxx
) can be deleted, which seems to be invalid logic.
Then you can use nested if instead of switch:
IF(
SELECTEDVALUE ( 'Table3'[Column0] )="Reve",
, CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text0" ) ),
IF(
SELECTEDVALUE ( 'Table3'[Column0] )="PPO",
CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text1" ) ),
IF(
SELECTEDVALUE ( 'Table3'[Column0] )="KO",
SUMX (FILTER ( 'Table2', 'Table2'[Column0] IN{ "Text2", "Text3","Text4","Text5","Text6"}),[Measure1]),
IF(
SELECTEDVALUE ( 'Table3'[Column0] )="PKO",
CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text7" ) ),
IF(
SELECTEDVALUE ( 'Table3'[Column0] )="EBIT",
CALCULATE ( [Measure1], FILTER ( 'Table2', 'Table2'[Column0] = "Text8" ) )
)
)
)
)
)
https://www.sqlservice.se/dax-quick-performance-tip-replace-switch-with-nested-ifs/
@martyska_sze See if this series of 2 blog articles helps:
Performance Tuning DAX - Part 1 - Microsoft Power BI Community
Performance Tuning DAX - Part 2 - Microsoft Power BI Community
There is a recent comment on Part 2 that demonstrates the potential value of optimizing your data model.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
A new Power BI DataViz World Championship is coming this June! Don't miss out on submitting your entry.
Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.
| User | Count |
|---|---|
| 53 | |
| 38 | |
| 33 | |
| 17 | |
| 17 |
| User | Count |
|---|---|
| 67 | |
| 63 | |
| 38 | |
| 34 | |
| 22 |