#DateForumTypeThreadPost
31608 Sep 2016 @ 12:08 PDTDatabaseReplyRules for Multi value compression 1. no MVC on PIs 2. Can have MVC on as many columns as desired, no such thing as too many columns. If there is a very large number of very large values, it is possible to exceed the total space li...
31507 Sep 2016 @ 07:59 PDTDatabaseReplyTeradata block storageCorrect. The row index list is just a set of offsets from the beginning of the block rather than a pointer but the effect is the same.
31406 Sep 2016 @ 07:25 PDTDatabaseReplyTeradata block storageThe rows in the data blocks are not sorted, they are just placed at the end of a heap of storage. The block contains an index list though which is ordered by the rowid. Likewise the blocks are not ...
31301 Sep 2016 @ 10:46 PDTDatabaseReplybetween condition make bad execution planWhat is the result of select count(*) from  tbla where TheDay < '2014-08-01'?  Trying to exclude the rows in tbla that cannot match from the product join.   Is the desir...
31201 Sep 2016 @ 09:04 PDTDatabaseReplybetween condition make bad execution planWhat described above requires the product join, every row of b compared to every row of a. This is going to be expensive for a medium table joined to a large table.  Are you sure the...
31131 Aug 2016 @ 12:46 PDTData ModelingReplyHow to transform an LDM to a PDM / Building block or Datamart?Either - depends on the bsuiness requirements. Generally the business requires the history of an account so it would be likely that Account is defined to save the expired rows. Events however often...
31030 Aug 2016 @ 05:50 PDTViewpointReplyIs there a way to control RESPONDING STATE of large extracts using TASM Exceptions?The result spool is not sent to BYNet. The BYNet is sent a reference to it where it was created by the last step of the plan. The result spool stays in the user's spool space until it is releas...
30928 Aug 2016 @ 06:42 PDTDatabaseReplyInserting into table using multiple common table expressions CTETeradata only supports one WITH. But within the one WITH it is possible to specify as many CTEs as desired.
30826 Aug 2016 @ 04:00 PDTDatabaseReplyPPIIt is generally better to start a new thread for a new question...   It is correct that a UPI cannot be defined when it does not contain the partitioning column. This is because enforcing th...
30726 Aug 2016 @ 09:48 PDTPrivate Forum
30626 Aug 2016 @ 09:37 PDTAnalyticsReplyQuery_Discussion_HELPPlease provide query and Explain.
30526 Aug 2016 @ 09:25 PDTDatabaseReplyTeradata: No more room in journal table for novaquality.journalsA SHOW table on the target table will showthe database location for the journals. It is possible the journals are in a different database that is out of space even if the database containing t...
30423 Aug 2016 @ 06:15 PDTDatabaseReplyCan we pass object as parameter in a macro ? Macro parameters are designed for constant substitution only. They can appear only in the places in a statment where a constant value can appear (and not every one of those either - like options in...
30318 Aug 2016 @ 08:12 PDTToolsReplyGetting export results in same date format as in Teradata table using BTEQPlease provide the SHOW TABLE. Need to see the data type and format for the column that contains the date.
30209 Aug 2016 @ 05:42 PDTGeneralReplyImprove performance on an analytical query , mulitple columns using same set of partitionsAs you can see, optimizer combines multiple ordered analytic functions with the same partition and order into a single execution.   Two different ones are required because there are two d...
30108 Aug 2016 @ 10:32 PDTGeneralReplyImprove performance on an analytical query , mulitple columns using same set of partitionsCan you provide explain please?
30004 Aug 2016 @ 10:49 PDTPrivate Forum
29903 Aug 2016 @ 05:59 PDTDatabaseReplyDistibution of Small table (rows less than number of amps) to single amp for Performance If the system has a very large number of AMPs, this technique can remove all-AMP steps from the query plans which access the table. If the queries that access the table are even medium size, the di...
29803 Aug 2016 @ 05:49 PDTDatabaseReplyKeeps returning zeros for decimal60 is an integer and when column1 is also an integer, integer division is performed which does not result in any fractional portion. Using 60. or casting either column1 or 60 to decimal will resolv...
29703 Aug 2016 @ 05:44 PDTToolsReplyExport with header using TPTIf the UNION form is used, it is a good idea to use UNION ALL so that the engine is not forced to do a duplicate elimination pass over the combination of the header row and the data.
29601 Aug 2016 @ 06:57 PDTDatabaseReplyMultiload The same way any data is loaded. A souce is decribed that contains the PI values to delete and multiload is told to load those PIs and then use them to delete rows.
29529 Jul 2016 @ 12:17 PDTPrivate Forum
29427 Jul 2016 @ 02:31 PDTDatabaseReplyMultiload There are two different forms of delete in Multiload which may be the source of confusion.   Multiload Delete Task accepts a DELETE statement with a where clause intended to bulk delete the ...
29319 Jul 2016 @ 09:10 PDTToolsReplyDetermine #sessions to set for load / exp jobsBetter yet, let WLM manage the number of sessions and only change the ones that really need exceptions via WLM rules. 
29219 Jul 2016 @ 09:07 PDTToolsReplyIn List Blues - How to make large In List Queries More Efficient What Teradata release is being used?   What does the explain plan look like for the 4500 class case?   On current releases, the Teradata optimizer should be doing this automatically. ...

Pages