Best Practices for Data Lab Performance

This article is intended for Seeq end users and should be referenced to optimize performance. If poor performance is experienced and the following best practices are followed, submit a support ticket.

Preserving Resources

Recommendations:

• Save and close all Jupyter notebook windows and tabs if they are not being used.

• Close the project by clicking the “Quit” button on the Seeq Data Lab home page which will terminate the Data Lab instance promptly and efficiently.

Reason:

The home page and notebook tabs left open when not being used will persist the Data Lab instance which consumes resources and may limit resources for other Data Lab project instances.

Scheduled Notebooks

Recommendation: Consider the work the notebook will do when picking a schedule. If it takes more than 5 minutes to run a notebook because it has a very large query or is utilizing advanced algorithms, do not schedule it to run every 5 minutes because that can place a constant load on Seeq which might impact other user activity. Schedule updates for when you need them. If there are times no one is using Seeq, those are great candidates for notebooks to run to minimize impact to others.