Archives of the TeradataForum

Message Posted: Fri, 23 Sep 2005 @ 20:23:33 GMT

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

Keith,

Our experiences are that dropping the partition is the fastest way (assuming all ranges defined and no secondary/hash/join indexes exist. Multiload delete is the next fastest then SQL delete is the slowest.

In Billion+ row (300+ gig tables) deleting about 1/12 of the data takes just 2 minutes or less (30 partitions). Multiload delete was 20 minutes. SQL delete was way too long as it used Transient Journal.

We are using V2R5.1

Thanks,

Bob Diehl

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