In most projects, the i2b2 ontology resides in a default database table called I2B2. If you have a large ontology, maybe consisting of concepts from different sources, it might be a good idea to split the ontology into multiple tables. That will simplify maintenance as you need to update only the table where the ontology has changed.
The table TABLE_ACCESS contains references to all other tables that are part of the ontology. See the example Boston demo database:
The column C_TABLE_NAME contains the names of tables (BIRN, ICD10_ICD9) in the i2b2 database schema that comprises the ontology. C_TABLE_CD contains a key which is used as a prefix for the path (C_FULLNAME) when i2b2 cells communicate.