Hi Team,
I have created one dataset , there has 2 columns (running & not running count) against the asset.
But when I use separate KPI widget for these Running & Not Running values. The first widget Running value came quickly, and the Not Running widget still refreshing and update later. In order to get those value from same dataset why these widget refreshing independently?.
I didn’t apply any filters on those widgets .And I refresh my dashboard through my application.
Hello @Marim,
Without knowing the exact dataset and the setup of the visuals it is hard to say for certain what is happening. In general, if the visualizations use different logic they will use different queries that can ran at different speeds on the underlying database / SPICE.
If you are running using direct query, you can check in your DB engine what queries are being executed and the time they take to return values, which should be aligned with the timings when you see your dashboards showing data.
Hope this helps!
Thanks @andres007
Here I used direct Query , In a single dataset I was getting these 2 columns values. If you see the below image if the Running count updated means the Not Running count also should update right?? I’m really doubtful about this
But KPI widget refreshing the Running Count first and Not Running Count takes some time to update
I joined this dataset with another one dataset.
I also ran this dataset separately without joins with another dataset, it took less than 6 sec in DB engine.
Can you help me understand?
Are the KPI visuals configured in the same way?
Maybe you can load a sample dataset and create the visuals in Arena so we can see what is going on?
Hello @Marim, I wanted to check in since we have not heard back from you with any further information. Is there a reason you are committed to direct query for this specific dataset? If load times are an issue or super important to you, I would always recommend utilizing SPICE and setting up incremental refreshes. This ensures your data doesn’t have to be queried from the database on load.
Also, from what I understand, the refresh runs independently for each visual when in direct query. Some visuals may take longer to load depending on the resource intensity. A delay is to be expected, especially if you have many visuals/calculated fields in the analysis.
Last thing to note. You said it took less than 6 seconds in the database engine to query, you should always anticipate longer query times when running from QuickSight. The database has to be queried, the data needs to be retrieved and stored into QuickSight, and then the visuals can load. Seeing longer times from an analysis on direct query is to be expected in this scenario.
Let me know if you have any remaining questions on this topic, and I am happy to guide you further. Thank you!
