#DateForumTypeThreadPost
16607 Jul 2015 @ 03:48 PDTDatabaseReplyQuery Tuning:PPI not being used in selectThat is because you are asking two different questions. In the first case you are asking for v_dsub to be qualified first before applying the rownumber function is applied. In the second case, the ...
16506 Jul 2015 @ 12:52 PDTHadoopReplyCan we recommend TDCH for Extract-Load processIt would be worthwhile to diagnose why the connections/jobs are failing.   An enhancement request is a great idea but it is not as simple as turning on checkpointing. A whole mechanism to re...
16404 Jul 2015 @ 07:18 PDTDatabaseReplyError from Insert Select statementDifferent data types? 
16329 Jun 2015 @ 12:18 PDTDatabaseReplyOptimizing Stats CollectionPerhaps you could clarify your next to last sentence: If a new event_dt(new records) is inserted in the table, it should be available in the system, without actually capturing the full st...
16227 Jun 2015 @ 04:22 PDTDatabaseReplyCreate tables for loading changing filesAnother alternative is to store some of the incoming record as a varchar and leave it up to the users to parse out the varying data at query time using the regexp functions. For instance if the fir...
16122 Jun 2015 @ 08:34 PDTDatabaseReplyNUSI when accessing data from a viewWhen you put "top 100 percent" in your view definition, you force the query to go through the sampling code and create a spool file. This eliminates the possibility of using the indexes o...
16019 Jun 2015 @ 07:48 PDTDatabaseReplyTables in one Database Vs Several DatabasesThere is no performance difference when joining/referencing across multiple databases.   The primary reason for using separate databases to organizae your tables is to logically group them f...
15928 Jan 2015 @ 12:10 PSTDatabaseReplyView refresh after collect statsWith a Teradata view, there is nothing to refresh. A view does not materialize the underlying data. It does not store data about the underlying tables. It is compiled on the fly as a query referenc...
15828 Jan 2015 @ 11:52 PSTDatabaseReplyDatabase space issuesLoad a sample of the data to ensure that your job is properly mapping the source field to the target PI. If it is loading nulls to the PI for instance, the job will be terribly skewed.
15727 Jan 2015 @ 07:15 PSTDatabaseReplySkew Factor & StatsSkew should not be considered important for any object smaller than a few data blocks per AMP. There is just not enough statistical distribution to get an even distribution. And skew for very small...
15631 Dec 2014 @ 07:29 PSTDatabaseReplywhere does TD store transient journal?You do not need to delete it. Teradata will clean it out for you automatically once you complete your transaction. 
15512 Dec 2014 @ 05:12 PSTGeneralReplyTeradata equivalent Query of SQL ServerCOALESCE is the SQL standard way to accomplish the job. Teradata has ZEROIFNULL and NVL as well. 
15422 Nov 2014 @ 01:10 PSTPrivate Forum
15322 Nov 2014 @ 01:09 PSTDatabaseReplyTemporary TablesIt is not necesssary to create a temprary table, that portion of your query can be written as a derived table in the FROM clause and then the rest of your logic can be applied between that derived ...
15222 Nov 2014 @ 01:05 PSTDatabaseReplyLock on DBC.Accessright table and other Create/drop/Alter statement in Block stateOn a platform like that, this ALTER should run very quickly. That suggestes to me that the ALTER itself is blocked for some period of time by some other piece(s) of work. You can investigate this i...
15121 Nov 2014 @ 09:45 PSTDatabaseReply issue with WITH clause and throing eroor like WITH IS NOT ALLOWED IN A SUBQUERY OR A VIEWWhat are you trying to accomplish with the WITH clause? The syntax above does not match any WITH syntax in Teradata SQL. 
15021 Nov 2014 @ 08:06 PSTDatabaseReplyConnecting Multiple Teradata ServersIf the tables are moderate size, the latest version of TDStudio has a drag and drop capability. Not appropriate for large table movement but very easy to use for small/medium on demand movement. &...
14918 Nov 2014 @ 10:23 PSTDatabaseReplyFallback optionA UDF in Teradata operates on the parameters provided to it rather than doing any local reads on their AMPs.   think in terms of creating an appropriate join and applying the UDF to the resu...
14818 Nov 2014 @ 10:12 PSTDatabaseReplySkew Factor of small tablesSee answers in: http://forums.teradata.com/forum/database/how-to-reduce-skew-factor-on-stage-tables-with-nopi
14718 Nov 2014 @ 05:56 PSTDatabaseReplyFallback optionCan you share what you are trying to accomplish?   fallback today can make a single additional copy of the data which will be placed on a different part of the configuration such that it it ...
14613 Nov 2014 @ 06:40 PSTDatabaseReplyLock on DBC.Accessright table and other Create/drop/Alter statement in Block stateMy second question was about what system and system size you are running this on. A VM on your laptop? A large production system? A small appliance?
14512 Nov 2014 @ 02:37 PSTDatabaseReplyHow to reduce skew factor on stage tables with NoPI?The PI values are put through a hash algorithm and a part of the hash is taken as the map to the amp for each row. When you have this small a sample, it is very likely that there will be amps with ...
14412 Nov 2014 @ 01:40 PSTDatabaseReplyHow to reduce skew factor on stage tables with NoPI?skew factor should be ignored for small tables. Until there are enough rows in a table to have several data blocks per AMP, the probabilities of the hash distribution are just not going to make the...
14312 Nov 2014 @ 03:51 PSTDatabaseReplyLock on DBC.Accessright table and other Create/drop/Alter statement in Block stateWhat Teradata database version and what kind of system are you running on?
14207 Nov 2014 @ 02:29 PSTDatabaseReplyIndex selectionLocks on indexes are separate from locks on the underlying rows. The referred to factors above are about secondary and join index selection rather than primary index selection. If a transaction upd...

Pages