Archives of the TeradataForum

Message Posted: Wed, 11 Apr 2007 @ 15:35:28 GMT

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

Thanks Bob for your thoughts and insight on the performance improvements to Multiload. Better performance always gets our attention.

Plus, it is really interesting to hear that INS/SEL is using block level I/O to the table. This does raise a question for me though: is the Transient Journal still being done a row at a time? If so, will this not slow down the merger operation to the point that there is no real performance gain?

I agree whole-heartedly with you that the speed to an empty table is amazing when the tables being used for the INS/SEL have the same PI. However, is that still the case when the target table has a different PI than the source table? Or is this a situation where Fastload would be a faster option?

Thanks in advance for the clarifications.

Best regards, Mike

Michael Larkins
Certified Teradata Master
Certified Teradata SQL Instructor

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