#DateWhereItemComment
2007 Mar 2010 @ 01:06 PSTcarrie's blogSome Rules are Meant to be Broken: Things that Don't Obey CPU Limits That is correct. DBQL will not indicate whether or not a query took longer to complete because it was being held back by a CPU limit.
1924 Feb 2010 @ 04:14 PSTcarrie's blogSome Rules are Meant to be Broken: Things that Don't Obey CPU Limits Imposing a CPU limit on an allocation group has no impact on TDWM delays that happen as a result of a throttle. Throttles can delay the time when a query begins execution. This shows up in DBQL. D...
1822 Feb 2010 @ 08:49 PSTDatabaseAll-AMP Random AMP Sampling? Sure, Why Not!Thank you for sharing your experiences. I would recommend to you to always follow the advice of the support center. The reason that turning on all-AMP sampling could impact query performance...
1716 Feb 2010 @ 12:54 PSTcarrie's blogWhat is the Estimated Processing Time? Response time is not the same thing as estimated processing time, although there is often a correlation. Response time is the difference in wall clock time between the time the query was submitt...
1615 Feb 2010 @ 03:26 PSTDatabaseAll-AMP Random AMP Sampling? Sure, Why Not!Hello Vincent, There are a couple of questions here, so let me address then one at a time. 1. USING SAMPLE - The percentage of rows read with the USING SAMPLE option will not increase for no...
1510 Feb 2010 @ 03:07 PSTcarrie's blogSimplifying AMP Worker Task Monitoring and Getting it RightI can try to clarify that for you... Messages arriving on the AMP that require an AMP worker task (AWT) to process the work they contain are assigned one of several different “types”. These ...
1404 Feb 2010 @ 10:02 PSTMarcio MouraWhen is the right time to refresh statistics? - Part IIHi Marcio, Thanks for sharing all this detail, this is good stuff! I have a question about your data dictionary statistics. When you look at that list of statistics that you collect on t...
1306 Jan 2010 @ 04:07 PSTcarrie's blogExpedite Your Tactical Queries, Whether You Think They Need It or NotThe ResUsageSAWT table shows you in-use counts for AMP worker tasks by work type. This table can be useful for understanding how many tasks are active for the expedited work types (usually work8 ...
1210 Dec 2009 @ 10:46 PSTcarrie's blogSimplifying AMP Worker Task Monitoring and Getting it RightAlmost, but not quite, so let me try to clear this up for you. (I think you meant "WorkOne", right?) "WorkOne=140" represents the sum of all AMP worker tasks in-use counts for each collect i...
1109 Dec 2009 @ 01:31 PSTcarrie's blogSimplifying AMP Worker Task Monitoring and Getting it RightThere are a couple of different examples of ResUsageSAWT output in the original text. Let’s consider them separately. The first example, under the heading The CollectIntervals Column, represe...
1003 Dec 2009 @ 08:34 PSTcarrie's blogControlling the Flow of Work in TeradataTerasuda, Both CPU time and CPU Disk Ratio are exceptions that can be defined on a workload in TASM. It would appear that your query exceeded one of those exceptions, which is why you got the ...
923 Nov 2009 @ 04:50 PSTDatabaseRemoving Multi-Column Statistics – A Process for Identification of Redundant Statistics Interesting approach. Thanks for posting, Dave. I am guessing here, but I think that the first column in the multicolumn stats where the distinct values did not change were both unique col...
813 Nov 2009 @ 09:54 PSTcarrie's blogControlling the Flow of Work in TeradataIf I am understanding you correctly, you are saying is that by the time you get an alert from Teradata Manager that your AMP worker task availability counts have reached some low threshold on at le...
709 Nov 2009 @ 01:56 PSTDatabaseCan We Speak Confidentially? Exposing Explain Confidence Levels Statistics collection has always been resource-intensive. A full table scan is done on the table and there are several aggregate operations performed for each statistic in order to calculate this ...
606 Nov 2009 @ 04:17 PSTcarrie's blogWhen Your Dirty Read is Kept WaitingKaram, You can delete all the rows in a partition by specifying a DELETE FROM ppi-table WHERE... command where the partitioning key value or range of partitioning key values describes rows conta...
508 Oct 2009 @ 05:13 PDTDatabaseStatistics Collection Recommendations for Teradata 12Thank you for your kind words, and for taking time to add a comment. The most common place I see unneeded statistics collected is with multicolumns stats. This is especially true if they are le...
430 Sep 2009 @ 08:12 PDTDatabaseStatistics Collection Recommendations for Teradata 12It sounds like this join you are describing is between 2 PPI tables, both partitioned identically, and the join is on both the PI and the partitioning columns of both tables. In this case the par...
313 Aug 2009 @ 08:25 PDTDatabaseStatistics Collection Recommendations for Teradata 12Good question. If the access of a single table is by primary index in an equality condition, statistics are not used and this will be a single-AMP operation, with no table scan. If the access...
217 Jul 2009 @ 09:49 PDTDatabaseSkew-Sensitivity in the Optimizer: New in Teradata 12I would not suggest that you attempt to force table duplication. It is better to leave join geography and join sequence decisions to the optimizer. But what you can do is make sure that you have ...
125 Jun 2009 @ 02:26 PDTDatabaseStatistics Collection Recommendations for Teradata 12I have incoporated the following response, and a few other minor updates, to the original article. Response: Correlated columns within a multicolumn statistic are columns where the value in...

Pages