The Sample node allows you to define Sampling at any point in an Analysis The specifications made in the Sample node will apply to the sheet displayed for the node itself and extend to any subsequent nodes in the Analysis' flow.
Note that the Sample node will interact with any Sampling settings made to the Analysis as a whole and at the Data Store Input level. Sampling in the Sample node is always applied after these other settings have taken effect.
For example, with the following combination of settings, a Sample node would display 100 records:
Analysis level settings: Max number of records to read: 5000, Sampling Rate (%): 5
= 250 records
Data Store Input settings: Max number of records to read: 250, Sampling Rate (%): 50
= 125 records
Sample node settings: Max number of records to read: 125, Sampling Rate (%): 80
= 100 records
The sample node can be particularly useful when part of an Analysis contains resource intensive computations and you want to quickly test these in a sheet.
Sample Node During Execution
Sample nodes can also be used to trim down the amount of data that passes through an Analysis when it is executed. This can be particularly useful if an input contains a large amount of data and you'd like to speed up execution time.