Archives of the TeradataForum

Message Posted: Thu, 06 Dec 2001 @ 06:04:29 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

For results return, Teradata ODBC driver transfers blocks the same way that CLI does. If result sets are being returned that are larger than a few rows, the ODBC buffer size parameter should be tuned to its maximum size, this will make a significant difference in result return performance.

For insert/update operations, neither ODBC or CLI batchthe updates automatically. That must bedone in a program. If Access is used to insert or update a large number of rows, access will send the rows a row at a time through ODBC which will circumvent all of Teradata's parallelism. If large numbers of rows are to be inserted or updated, the utilizing the appropriate Teradata utility is indicated.

It is Teradata's strategic intent that ODBC and CLI perform equivalently. We would always recommend ODBC as the appropriate open interface to use for new applications on platforms other than the mainframe.

	<Prev	Next>		<<First	<Prev	Next>	Last>>