Impact Analysis is the process of analyzing and identifying the impact a data object has on other objects in the data catalog.
Impact analysis has a dependency on lineage being available for all the source data systems in OvalEdge. Any object missed out in the lineage building process will be missed out by the impact analysis.
The object for which impact analysis is being done, is referred to as the Affected Object.
Impact analysis can be performed under any one of the following scopes:
- UPSTREAM: This scope will identify the objects that are part of the backward or upstream lineage for the affected object. In the lineage graph, these objects will be on the left side of the affected object.
- DOWNSTREAM: This scope will identify the objects that are part of the forward or downstream lineage for the affected object. In the lineage graph, these objects will be on the right side of the affected object.
- DOWNSTREAM with Associations: This includes the associated objects of the datasets, which are part of the backward lineage of the affected object, within the scope of impact analysis. This can potentially increase the impacted objects geometrically.
- UPSTREAM with Associations: This includes the associated objects of the datasets, which are part of the forward lineage of the affected object, within the scope of impact analysis. This can potentially increase the impacted objects geometrically.
Another important element defining the scope of impact analysis is ‘Impact Level’. This is the level of separation of the impacted object from the affected object.
- When specified as an input at the time of doing impact analysis, it limits the identification of impacted objects to level specified. For example, if the impact level is set to 3, then impacted objects that are more than 3 levels upstream or downstream are ignored.
- When used for analyzing the output of an impact analysis, it helps in filtering the impacted objects to the level specified.
OvalEdge Impact Analysis can be performed on:
- Tables
- Files
- Reports
- Table columns
- File columns
- Report columns
The impact analysis results are explained on a sample lineage graph diagram below.
It shows the affected object, at level 0, and its downstream impacted objects at various levels.
It is important to note that only data storing objects like tables, files, reports, columns have an impact on the impact level in a lineage path. Dataset query objects in a lineage path do not affect the impact level as the do not store data within them.