Archives of the TeradataForum

Message Posted: Thu, 26 Mar 2009 @ 13:01:06 GMT

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

Mark,

You can do this, but you need to recover each of the cluster files as all-amp COPY jobs (i.e. you can't run the restore to specific clusters).

www.teradataforum.com/teradata/20040209_202017.htm

To minimise the serial limitations around this process, you need to ensure that you're archiving data at table level, instead of at database level (you can put together a quick script to write the archive code for you - lot quicker than coding this yourself!). Multiple restorations can then run simultaneously, with the COPY jobs running 'staggered' around table locks, i.e. job B can load its data to DatabaseA.TableA, as soon as Job A moves on to DatabaseA.TableB, instead of DatabaseB.TableA.

Hope this helps,

Thanks,

Mark

Mark Dodgson
Data Architecture, Barclays Bank Plc (GRCB Technology)

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