I've previously asked this in the developer forum, but it's a little quiet over there, so I thought I'd try here too -- hope that's ok...
I have a streaming dataset with a DateTime field. The streaming dataset is keeping historical data. When I create a report to put on a dashboard using the DateTime field, I do not have any date hierarchy capabilities. I am therefore not able to see my data by day/month, only by time.
Whilst it's great to see the data realtime, some departments only want to see that data summarised over day/week/month not by minute. I'm truly hoping that I'm doing something wrong rather than a fundamental piece of functionality being missing.
Any thoughts greatly received.
I don't have a streaming data set but maybe this will be the same: in Dataset - starting with a blank canvas, select your date field from the right Fields pane so it shows up in the canvas
then in the Visualization pane, look at it's drop down and see if Date Hierarchy is checked.......
So, to be clear, this is a streaming dataset with the field set as a DateTime field. When shown on a live tile, it shows the data updating every 30 seconds (as that is how often I'm updating the data), so the field is recognised ok as a date time field. It's just that when I put the field onto a normal report (rather than realtime), date hierarchy is not an option so that I can view average stats over a day/week/month/year etc.
I hope that's clearer.
I'm facing the same issue. I can only use date hierarchy when I use a normal (not streaming) data source.
As a workaround, I pushed the same JSON data to Azure Event Hub, then with a Stream Analytics job copied it into a DocumentDB database.
The DocumentDB database can be used as source with Power BI Desktop. If you publish the report to Power BI online service, data can be refreshed with a schedule (pro feature) automatically from Azure DocumentDB. Since the source is not a stream anymore, it is not useful to show real time data on a dashboard, but you can build better reports.
However, if you keep pushing the same data (same JSON string) straight to Power BI too, you can display that data on a dashboard in a real-time way.
I am having the same issue. PBI Desktop doesn't seem to support the data heirarchy feature on live streaming data. Some of the visualizations don't have the data heirarchy feature at all, even with the data type set to DateTime.
Hi, I am facing sort of same issue.
When "Historic Data Analysis" tab is checked, it does not take timestamp as Datetime, rather accepts it if i change it to Text.
But unfortunately I am unable to use the date hierarchy while creating reports from the same since it recognizes it as just a string.
I'm guessing this is a missing feature for streaming datasets. I've tried to accomplish your exact solution for the same reasons but failed to create the hierarchies. Adding new measures for dateparts and using those as drill down layers might work for some visuals, but to be honest I'm not about to spend that time.
The main reason behind this issue might be that creating detailed aggregated reports may not be what PBI team is aiming with streaming datasets. Let's agree on the best way they work is when you need to monitor frequent data, like sensor streams. I'm leaning towards seperating the datasets for respective purposes: real-time and warm data.
@dparkinson have you found any workaround to this?