Release 7.1.0.0 includes the following changes to the ODBC,
JDBC, and OLE DB data connectivity clients.
- The ODBC and JDBC drivers, and the OLE DB provider, now support
the ability to connect to a schema within a database that has multiple
schemas defined. For more information, see the IBM Netezza ODBC, JDBC, and OLE
DB Installation and Configuration Guide
- The ODBC driver now offers the ability to bypass the data type
conversions that it performs during SELECT queries. Skipping the data
type checks and conversion can significantly reduce the processing
time for the data retrieve and improve the performance for the SELECT()
queries issued by applications that use the driver. With this change,
users can set an option called FastSelect in the odbc.ini file
or in the connection string to cause the DSN connection to skip the
data conversion processing for the data types read from the NPS® system
and which will be processed by the ODBC application. The ODBC application
is responsible for ensuring that the retrieved data types match the
expected return values, or the query will fail with data type errors.
- The JDBC driver has a new connection setting Ignore UpdateCount
that you can use to override the JDBC specification for getUpdateCount(),
which is an int value. If you use CREATE EXTERNAL TABLE to unload
data and the row count exceeds the maximum value of an int (two billion),
the command fails with the error NzSQLException: The update
count exceeded Integer.MAX_VALUE. This setting can be helpful
in cases where you are exporting data from an NPS table
to an IBM® PureData™ System for Hadoop appliance.
- If you use JDBC to connect to a Netezza® host that uses
SP 800-131a cryptography support, you must install and use Java Runtime
Environment 1.7 to support the enhanced cryptography connections.
IBM PureData System for Analytics, Version 7.1