5.1.5 Considerations for using transformational functionality Some things to consider when using InfoSphere CDC transformational functionality are: Depending on the InfoSphere CDC engine you have installed, some column functions might not be available. Column function syntax may differ between platforms. For example, the syntax for %GETCOL is different for the IBM i platform than the one used by DB2 on Linux, UNIX, or Windows. Names for column functions are not case-sensitive. For some column functions, large object (LOB) columns cannot be specified as a function parameter. Character literals can be specified in their internal numeric representation, as parameters for column functions. To accomplish this task, use the double-angled bracket notation (<< >>). This notation allows both printable and non-printable characters to be used. Specifying character values as decimal integers can represent either American Standard Code for Information Interchange (ASCII) or Extended Binary Coded Decimal Interchange Code (EBCDIC) characters. 5.2 Replication modes InfoSphere CDC provides three modes of replication: Refresh Continuous Scheduled End (Net Change) These modes provide flexibility about how and when to replicate. Refresh replication provides bulk copy functionality. Continuous and Scheduled End replication provide log based and mirroring functionality. 5.2.1 Refresh The InfoSphere CDC refresh operation is the replication mode used to capture a complete copy of the data in the source table and transfer that data to the target table. Each refresh applies all features of replication (such as row and column filtering, column mappings, and so on) during the transfer. Chapter 5. InfoSphere CDC features and functionality 83