To improve performance of the Tableau Dashboards we can implement the following approaches.
. Push complex calculation to the Database.
. Use materialized views in the Database.
. Create indexes in the Database.
. Use extracts along with the filter out unwanted data.
. For each filter a separate query is created by Tableau, so use less number of filters.
. Data cleansing at database level is better than at Tableau.
. Boolean Data type are faster than String & Date calculations, if possible use Booleans in the calculations .
. If we want to show multiple measure values then use parameters to dynamically select the
measure values, it improves the performance over showing all measures at a time.
. Adding context filters to the workbook will degrade performance,
since if any change takes place in the database, it will update the views in the tableau by
computing the calculations again can lead to reduce the performance of Dashboard.
. Enable show relevant values for Quick Filters.
. If we are creating very complex calculations, First create sub calculations then call them into main calculations.
. Applying filters on multiple sheets generate multiple queries, so apply wherever it is required.
. Use Continuous date filters improves performance over Discrete Date filters.
. Remove unused databases, fields, sheets, dashboards, Filters, Calculations.
. Grouping functionality loads the entire data, for performance points of view Use Case statements or sets instead of groups.
. Native Database drivers than ODBC.
. Avoid using Custom SQL.
. Case is faster than if then else.
. Convert the Date field from raw string to the string or number.
. Limit the number of filters showing on the view for space constraint and for performance improvement.
. Reduce the number of marks on the views.
. Turn off Automatic updates, like update filter values, Pass Auto Updates.
. Better to use Include filter than exclude. And use continuous data filters rather than discrete.
No comments :
Post a Comment