How to improve the schedule exporting performance for dashboards
Handling Schedules for Multiple Users in Dashboards with User Filters
When configuring schedules for multiple users in a dashboard with user filters (row level security), it is essential to understand the impact on server performance and potential delays in dashboard rendering. This article will discuss the behavior of schedules in such scenarios and provide recommendations for minimizing performance issues.
Server Performance Impact
If you have 25 dashboards without a user-based filter configuration, for each dashboard, one request will be made, and the same exported dashboard will be sent to all 40 users. In this case, there will not be much impact on the server.
However, if you have a user-based filter configured for each dashboard, 40 requests will be created for each dashboard. This means that 40 x 25 (1000) requests will be processed on the server simultaneously. During this time, if you have active users accessing the dashboards, there might be a performance delay in the dashboard rendering part for the user.
This delay depends on multiple variables, such as:
- The number of data used in the data source
- The number of complex expressions created in the dashboard
- The processing power of the server that holds the data
Additionally, each SMTP service has a limit on the number of emails that can be sent in a given time.
Recommendations for Scheduling
To reduce the error ratio due to the above-mentioned variables, we recommend not scheduling all the dashboards at the same time. Instead, you can schedule the dashboards with a few minutes delay between each. This approach will help minimize performance issues and ensure a smoother experience for users accessing the dashboards.
In summary, when configuring schedules for multiple users in a dashboard with user filters, it is crucial to consider the impact on server performance and potential delays in dashboard rendering. By scheduling the dashboards with a few minutes delay between each, you can minimize performance issues and provide a better experience for users.
Additional References
