Forum Discussion
RANKX By Group In Measure for Push Dataset
kbsmx we are going in a circle, the solution (measure) I provided check the max date of each site, if there are late arrival data, it will take the most recent date of each site, it not max date across all the site but the individual site, not sure what I'm missing here. I think that's pretty much what I understood and maybe I'm not getting your point here, you have to throw example data and show what is not working otherwise we will keep on going back and forth.
parry2k - I hear you, but we're not going in a circle. It's not behaving the way you describe. Do you have a push dataset you can test this against? Perhaps you are testing this with a local data model vs a push dataset which has very different behavior and limitations with some calculations.
Here's what I'm using based on what was provided by MFelix :
Latest Site Refresh Date = MAXX(SUMMARIZE(ALLSELECTED(RealTimeData), RealTimeData[Site], RealTimeData[Status Date]), RealTimeData[Status Date])
Is Latest Site Data = IF(MAX(RealTimeData[Status Date]) = [Latest Site Refresh Date], 1, 0)
Map visual filter set to:
parry2k MFelix - Any other ideas? This is affecting two different reports using two different push datasets. Because both datasets receive new data every 5 minutes, and it takes approximately 1 minute for the new rows to post, there's about a 20% chance every 5 minutes that a user will see the partial rendering which we need to avoid. I appreciate your help and understand that this is a difficult nut to crack.
HI kbsmx,
I think these should more relate to the Dax formula calculation on 'live' mode data source. These type of data table that update frequently, they may affect the dax formula calculations.
In my opinion, I'd like to suggest you direct calculate and append these results into the source data and push with these requests at the same time. It should reduce the effect of Dax calculation that hosts on frequent updates data sources.Regards,
Xiaoxin ShengAnonymous Unfortunately, that's not possible. You can't precalculate the results for this type of filter as you can't query the data in Power BI to produce such a calculation nor can you update existing rows in a push dataset. This filter *must* be done in DAX based on the current push dataset API limitations.