BIND command reads the SQL statements from the DBRM and produces a mechanism to access data(access path), in an efficient manner, as directed by the SQL statements being bound.
The access path is determined by DB2 during the bind process. This determines what indexes will be used.
DB2 Access path is (in most cases) the path which results in the most optimized execution of an SQL query. Whenever any Static SQL (Embedded SQL) is bound to a package/plan, the DB2 optimizer creates the most efficient (optimized) way to access the data from the Query using the Predicates Used in the QUERY (AND,OR, GROUP BY, ORDER BY, >=, IN, IS NULL,etc.) and also the information from DB2 Catalog of available indexes on the tables used in the SQL and also using the table statistics (no. of rows in tablespace, no. of distinct values of a column,etc.) . The result is an access path which decides which predicates to filter first (indexable, stage1 or stage2). The Access Path is stored in the Package/Plan in the DB2 Directory which is not Queryable. The DB2 Catalog merely stores information about Access Path created during BIND in the OWNER.PLAN_TABLE Catalog table.