Forum Discussion

JS-cubelogic's avatar
JS-cubelogic
Advocate II
8 years ago

Table Visualization results uncontrolled SSAS memory growth

I created a report with a table visualization running against an SSAS Live Connection. The table contained 5 dimension values and 2 measures. There were many rows of data > 1000 in some cases. The re...
Show and Tell
v-qiuyu-msft's avatar
v-qiuyu-msft
Community Support
8 years ago

Hi JS-cubelogic,

 

I find the same issue reported before, and you can refer to below information: 

 

SSAS 2016 changed behavior to avoid possible incorrect results for multi-level nested operations in DAXMD, as a result requires that we materialize some of the intermediate results which can generate very large datasets. 

 

The behavior seen in customer is by design, so SSAS and Power BI team are working on VSTS 9229662 to change Power BI queries in order to user SuperDAX, which will show better performance.

 

The workaround is continue to use SSAS 2014 or simplify queries to avoid pattern that consumes a lot of resources in SSAS 2016.

 

I will send a email to consult the status about VSTS 9229662. Once I get information, will update here. 

 

Best Regards,
Qiuyun Yu