In Savant, a run is the process of executing your analysis from start to finish. It allows you to see the results of your configured tools and transformations on your data. Runs help you validate your data processing steps, ensuring everything works as expected.

When configuring tools and interacting with data on the canvas, Savant provides a default sample of 1,000 rows. This feature lets you quickly visualize your analysis and refine your configuration before running the full dataset. Sampling on the canvas ensures a smooth and efficient experience when working with large datasets.

If you need to work with more data than the default sample, you can press the "Analyze" button. This action processes up to 100 MB of data, allowing you to validate your analysis on a larger subset without running the entire dataset. The Analyze button strikes a balance between speed and accuracy, helping you fine-tune your workflow efficiently.

Tests: Tests in Savant run your entire workflow on the full dataset but do not write any outputs to destinations. They are perfect for validating your analysis, ensuring your tools and configurations function as intended without affecting downstream processes.

Runs: Runs execute your workflow and write the results to specified destinations. They are the final step in your analysis and are used when you're confident in your workflow and ready to generate outputs for further use or distribution.