Traceability

TM supports two distinctive types of traceability - namely 'Instance Data Traceability' and 'Metadata Traceability'.

Metadata Traceability allows the transform designer quickly to work out which sources are linked to which targets. This can be done from the target or from the source perspective. From the source perspective we can work out which targets will be impacted by a change of particular source attributes - this is impact analysis; however, it is often just as useful to consider the problem the target perspective - this is basically an audit analysis, or data lineage. A fascinating feature of seeing is that audit analysis may be chained across a series of projects - allowing one to trace back the ultimate source(s) of particular data values.

Instance traceability, on the other hand, is much more a tool used by developers when developing transforms. It allows the TM Design Tool to display which sources and target instances are read, linked or constructed as the transform executes. Two types of tracing are available - XML to XML, and database to database. The XML to XML tracing is particularly useful when the XML documents are large - the exact location of the relevant elements is automatically displayed. Anyone who has worked with larger or more complex XML documents will know how much time this saves. Database to database tracing is even more powerful. After a transformation run TM is able to display step-by-step only those elements, linked by foreign and primary keys which participated in the transform. This is particularly useful because the target database is already completely empty. Anybody who has spent hours with TOAD or SQL worksheet following chains of foreign primary key relationships will appreciate the value of this automatic navigation.