Archives of the TeradataForum

Message Posted: Fri, 10 Oct 2008 @ 15:21:57 GMT

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

Archive jobs will use CPU time on the MVS machine and in Teradata. To capture the CPU time used in MVS, one option is to scrape the job log. It will say something like "STEP ARC01 ENDED - TCB TIME 0:00:03.123 SRB TIME 0:00:02.234"; you need to add the TCB and SRB time to get the total CPU. Your MVS support can undoubtedly suggest more sophisticated methods.

On Teradata, an archive uses multiple sessions, so you would need to find all sessions associated with a job and add up the CPU time from DBC.AmpUsage. This should be easy enough if you have distinctive account strings for archives.

Space could be a little more challenging. The archive log reports the number of bytes archived for tables, but as I recall it doesn't show space for views and macros. The tape management system on MVS will report the number of blocks in each data set on tape, so maybe that would be a better source.

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