Archives of the TeradataForum

Message Posted: Tue, 24 Jul 2012 @ 15:48:19 GMT

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

For example ...

     REPLACE PROCEDURE LOAD_COMMENTS_P
     (
     IN COMMENT_DT DATE,
     IN LINK_ID CHAR(7),
     IN BUL_ID_COMMENT DECIMAL(5,0),
     IN METRIC_TYPE_ID VARCHAR(255),
     IN COMMENT_TX VARCHAR(255)
     ) DYNAMIC RESULT SETS 1

do lots of dml and sql statements ... then ...

     BEGIN DECLARE cur1 CURSOR WITH RETURN ONLY FOR
     SELECT ReturnText AS return_row FROM LOAD_COMMENTS;
     OPEN cur1;
     END;
     END RETURN_ROW;

     END;

Now that we've got our BI Tool to read the data returned from a SP, the question would have to be : with Dynamic Result Sets (and hence technically a Cursor), would it be slow for lots of data?

We've had this conversation in past wrt For Loops and Cursors but, given that this is a sort of short-cut syntax, does Teradata use All Amps to satisfy this Cursor, or is it still inherently a one Amp process?

Put another way, is this a robust and performant way of gathering data together - which may require different sql statements to be run, depending upon the input parameters - and then presenting back to the EndUser Tool?

Thanks for any informative comments.

Regards

David Clough
Senior BI Database Designer

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