Consider adding the following types of attributes to the snapshot tables:
-
Primary keys: Primary keys are accessed often.
-
Join attributes in linked repositories.
-
Attributes in Preferences that are used frequently: In a Repository View, if you have a Preference that doesn't have many columns, that you will use frequently, and that you want to display as quickly as possible, consider putting all its attributes in the snapshot table.
-
Attributes that you want to be able to sort: When you view a repository in a Repository View using the Grid View, only attributes that appear in the snapshot table can be sorted. Attribute columns can be filtered whether or not the attributes are in the snapshot table.
-
Attributes you want to range search: To be able to search for a range of values on attributes of datatypes integer or date, you must add them to the snapshot table.
-
Attributes you plan on accessing through SQL: Attribute values can be read from the snapshot table using SQL.
-
Attributes to be used in a bulk callout: If you want to access an attribute in a bulk callout, it must be in the snapshot table.
-
Attributes included in a Scheduled Export of type View: If an attribute is included in a Scheduled Export of type View, unless the attribute is in the snapshot table, it must be extracted and parsed from the XML.
-
Attributes accessed from workflow: If a workflow accesses an attribute value, unless the attribute is in the snapshot table, it must be extracted and parsed from the XML.