This method considers the various factors
of the specified output size (in blocks),
and returns the highest factor that is less than
the number of available buffers.
This method considers the various roots
of the specified output size (in blocks),
and returns the highest root that is less than
the number of available buffers.
Estimates the number of block accesses to compute the
index selection, which is the same as the
index traversal cost plus the number of matching data records.