#DateForumTypeThreadPost
29114 Jul 2016 @ 06:17 PDTDatabaseReplyAvoid all-row scan in driving table using NUSIIt is not ever possible to test NUSI usage on small data. If the data is only a block or two on each AMP, it is always cheaper to read the data blocks directly than to read the index structure...
29013 Jul 2016 @ 07:42 PDTDatabaseReply3899 Internal error in teradata SQL parserIt is really best to open a new incident with Teradata support. It is not possible to determine if this issue is the same as the one in the original thread without analysis of the details.
28913 Jul 2016 @ 07:36 PDTDatabaseReplyextending MLPPI tablesBut it is not neccesary to drop and reload the table. Create the new table definition with the new partitioning, then use INSERT SELECT to move the data from the current table to the new one. INSER...
28812 Jul 2016 @ 06:14 PDTDatabaseReplyStoring History DAta into tablesMulti-value compression will NOT slow down access time, if anything it will improve performance by requiring fewer IOs. 
28711 Jul 2016 @ 05:25 PDTDatabaseReplyDifference between event code and crash code3610 is a generic code that says something happened that the Parsing Engine couldn't deal with. The crash code is additional information that tells what area of the Parssing Engine had the issu...
28601 Jul 2016 @ 03:51 PDTGeneralReplyTeradata to AWS(Redshift)Or better yet just use Teradata on AWS to run your work there!
28528 Jun 2016 @ 08:45 PDTDatabaseReplyCOLUMNAR TABLE- SELECT TOP N* FROM COLUMNAR TABLE NOT GIVING OUTPUTWas it possible to create an incident for this?   We did some testing and were unable to reproduce the issue.  If there is an incident, please share the incident #. If not, perhaps some...
28421 Jun 2016 @ 04:54 PDTAnalyticsReplyCPU disk ratio error from 'like' statementsPlease share the error code and message being received? It sounds like a workload management rule that the system/performance administrators have put in place. There are no system limitations aroun...
28321 Jun 2016 @ 12:48 PDTDatabaseReplyCOLUMNAR TABLE- SELECT TOP N* FROM COLUMNAR TABLE NOT GIVING OUTPUTIf you or someone at you site has access to Teradata support, please open an incident for this.
28217 Jun 2016 @ 12:21 PDTDatabaseReplyRecursive Query Vs UNIONNot sure what "roll it up" means since the result doesn't obviously include an aggregate of any of the columns. The previous answer is the beginning but becomes a subquery: SELECT t1...
28117 Jun 2016 @ 12:05 PDTDatabaseReplyRecursive Query Vs UNIONIm not totally clear on wht you want but I think it is a list of unique table1.a values where any of the table1.d values is less than zero...   SELECT a FROM table1 GROUP BY a...
28016 Jun 2016 @ 05:18 PDTDatabaseReplyCopying trillions of data to new tableIt would be good to see the explain.   The number/granularity of partitions does not affect the cost of executing this process (and converting the granularity of the partitioning would be v...
27916 Jun 2016 @ 05:05 PDTDatabaseReplyAlter table on huge tableI assume that @samir means that they usually do: INSERT SELECT old table to new table Rename old table to save table rename new table to old table drop save table when satisfied that result is ...
27808 Jun 2016 @ 04:00 PDTGeneralReplyOrder of execution in Qualify clauseThe order of evaluation within the Qualify clause is not important because all conditions in qualify are executed after all of the ordered analytics have been calculated.    If if what ...
27708 Jun 2016 @ 07:03 PDTToolsReplyError locating in SQL Assistant 15It would help if a description of the errors encountered was provided.
27607 Jun 2016 @ 10:30 PDTDatabaseReplyWhen a new node is added in teradata then the how the remapping will take place?The hash stays the same, it is not recalculated. A subset of the bits of the hash are designated as the "hash bucket". In Teradata there is a map that assigns buckets to AMPs. When new AM...
27503 Jun 2016 @ 02:16 PDTDatabaseReplyUDFs and Granting of Access RightsUsername does not belong in single quotes. If it contains any special characters, it should be enclosed in double quotes or if no special characters then simply say the username.
27403 Jun 2016 @ 10:43 PDTToolsReplyTPT API C program usageTPT EXPORT is going to be better performance than a coded solution not using the export operator capability.   Generally performance of export is gated by client CPU, client write target(eg ...
27302 Jun 2016 @ 11:56 PDTDatabaseReplySpool space and MVCWhen you do a select, the result is formatted in it's final retrieval form in the final spool file. This means that dates, numbers, timestamps,... are all translated into character formatting, ...
27202 Jun 2016 @ 10:06 PDTDatabaseReplyTable ID and Table NameThis is why there is a DBC.Tables view. The views in DBC do all the joins for you so you don't have to know the details of linking tableids, databaseids,... The views should always be used rath...
27131 May 2016 @ 07:26 PDTDatabaseReplyCREATE TABLE Failed [3737] - NAME REQUIRES MORE THAN 30 bytes IN LATIN INTERNAL FORMThe message about latin internal form is a reference to the fact that names of objects may now contain non-latin characters. If characters are used that require two bytes to store, then the number ...
27026 May 2016 @ 10:29 PDTDatabaseReplyPartition by Range_NWhat is the goal for partitioning this column? With the UPI, any query WHERE column1='xx' will already be a single AMP, single row query.
26926 May 2016 @ 10:00 PDTDatabaseReplyLeft join and filter NULL doesnt work!Please provide the query, results, explain and database release.
26819 May 2016 @ 06:56 PDTDatabaseReplyHelp tables shows truncated column namesUnfortunately we had to maintain compatibility for applications which were coded to use these fields in the past. So we kept the existing fields the same size as they used to be and added new field...
26716 May 2016 @ 07:09 PDTViewpointReplyIs there a way to control RESPONDING STATE of large extracts using TASM Exceptions?What is the goal?  What are you trying to manage?

Pages