Improved fast access to individual rows

DB2® 10 introduces a new access path for faster access to individual rows by using a fully qualified key rather than a traditional index. This access path is called hash access.

Hash access can reduce CPU time and potentially I/O wait time. When DB2 selects a hash access path, only one I/O is required to retrieve the row from the table, which reduces CPU usage. A table space with hash organization might require additional storage to minimize the number of overflow records.

Hash access also can eliminate the need for table space scans and index scans for access to a single unique key.

Hash access is efficient for certain types of tables and queries, such as:

• Queries that use equal predicates to access a single row on a table
• Tables of a predictable and reasonably static size. To create an effective hash algorithm, DB2 needs a close estimate of the volume of data that is expected to be in a table before the table is populated.