[1] In analogy with relational databases, a super column family is something like a "view" on a number of tables.
[2] It is useful when making a data model to have some kind of a view on a number of tables.
Using a super column family is something similar to that in distributed data stores.
There are, however, no "joins" between the "tables", as data stores like Apache Cassandra are non-relational.
Similar to the standard column family, sorting is defined by an attribute.