The Dataset Splitter divides the current Data Table into sub-tables based on the values which appear in a specified column. For example, consider the following Data Table:
If we split the data over the "Group" column, this will divide the Table into four sub-tables, illustrated below:
Notice that each sub-table now has a unique "Group".
What happens to these sub-tables? Each one is passed to the next Task Element in a separate Execution Path so that it can be handled individually. For example it can be turned into an HTML Table, or used as the basis of subsequent queries, etc - see Iterative Execution Paths.
To join the sub tables back together into a single table use the Merge Data Node.