Archives of the TeradataForum
Message Posted: Thu, 14 Apr 2011 @ 11:45:26 GMT
<-- Anonymously Posted: Tuesday, April 12, 2011 15:39 -->
We use the ASE variables &S&D&H. I have a couple of questions regarding HOW the ACCTG buckets are populated and WHEN they are populated, when using &S&D&H.
For example, suppose a user whose accountstring ASE variables are &S&D&H starts a query at 2:30am, and the query completes at 4:30am.
Question #1. HOW do the Acctg buckets get updated? Would appropriate portions of the CPU and IO for this query go into three separate buckets that is buckets for hour 02 , hour 03 , and hour 04 ? Or would everything go into 1 bucket? If one bucket, is it the bucket for hour 02 (the start of the query), or hour 04 (the end of the query)?
Question #2. WHEN does the Acctg table get updated? Does it get updated AFTER the query completes at 4:30am? If the CPU and IO are split across multiple buckets (see preceding question), do all three buckets get updated after the query completes, or do the hour 02 and hour 03 buckets get updated at the end of their hour?
Question #3. When using &S&D&H, is the accountstring expanded at the start of the session, or the start of each request within the session?
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|