The new Data Explorer user interface is available in Public Preview and includes new features that weren't present in the old interface. To learn more, see Preview Features.
Sampling in Atlas is the selection of a subset of documents from a collection for analysis. Analyzing a sample set of data is a common statistical analysis technique; the results of analyzing a sample set tend to be similar to the results of analyzing an entire data set.
Atlas uses sampling for efficiency. Generally, standard sample sets can be selected and analyzed in a few seconds. Analyzing large samples or entire collections demands significantly more time and processing power.
Sampling Method
Atlas samples 1,000 documents from a collection using the $sample operator via the aggregation pipeline. This provides efficient, random sampling without replacement over an entire collection, or over the subset of documents specified by a query.