Using a Conditional Split
Data Transformation
This chapter shows you how to create a package that can split data based on a given
set of criteria into separate flows, a conditional split of data. You will also learn how
to use the in-memory data flow destination, the Recordset Destination.
Often, you want to separate out data meeting certain established criteria and direct
them to different destinations. The data splitting may be needed for any number
of reasons, to focus on a business objective: weed out data that are suspicious; data
that does not follow some standard pattern; identify missing data, etc. The data flow
transformation, Conditional Split Transformation is ideally suited for such
an operation.
Hands-On Exercise: Splitting Data
Retrieved from a SQL Server
In order to follow the steps as indicated, you will need a data flow task that connects
to a data source, and destinations to which the separated flows can be directed into,
which includes a destination for the default flow that does not meet the criteria
established for data splitting. You will also need to add and configure an in-memory
ADO recordset destination, the Recordset Destination component to visualize
the data.
The following are the major steps of this exercise:
Create a BI Project and add a Data Flow Task. Add and configure the
DataReader Source to pull data from the Local SQL Server.
Add a Conditional Split Transformation.
??? ???
Using a Conditional Split Data Transformation
[ 136 ]
Establish a path to Connect DataReader Source with the Conditional Split
Data Transformation.
Pages:
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107