#DateForumTypeThreadPost
280311 Jun 2014 @ 02:32 PDTDatabaseReplyDBQLFlushRate under 60 secI don't know about the exact impact of a more frequent flush, but 600 seconds are just a recommendation and the minimum value is 1 second, so 30 seconds should be ok. There size of the DBQL sh...
280211 Jun 2014 @ 02:08 PDTDatabaseReplyHow to read VARBYTE@HateOra: Define "readable format" :-) It's binary, so you (or you application) should know what it's supposed to be, e.g. JPEG, MP3, etc.
280111 Jun 2014 @ 02:05 PDTDatabaseReplyneed syntax for for last 2 years dataThere were already multile threads on first/last day of month. first day = dt - (extract(day from dt) -1)  last day = first day of the following month - 1 (or your TD version supports LAST_D...
280007 Jun 2014 @ 10:27 PDTDatabaseReplyneed syntax for for last 2 years dataDon't use the INTERVAL syntax as it will fail on 2016-02-29, better use ADD_MONTHS instead.
279907 Jun 2014 @ 10:24 PDTGeneralReplySpool errorDid you check dbc.QryLogStepsV which step failed? Could you show the DDLs? I would suggest to switch to a MERGE instead, this might be more efficient.
279807 Jun 2014 @ 10:18 PDTDatabaseReplyGroup By an Partition by in a single queryYour query doesn't make sense, as the combination of COL1,COL2,COL3 is unique, so SUM(COL2) OVER (PARTITION BY COL1,COL2,COL3) returns the same as a simple COL2. Of...
279730 May 2014 @ 09:24 PDTToolsReplyBTEQ - show command loses formattingThis might be due to different linebreaks on Unix/windows. Did you use SHOW TABLEs? Did you run BTEQ on Linux and open the DDLs on Windows? e.g. WordPad can deal with Unix linebreaks, but not No...
279629 May 2014 @ 08:44 PDTDatabaseReplyoptimization of query"Additional details" means DDL, actual query and Explain...
279528 May 2014 @ 12:00 PDTDatabaseReplyCollecting Sample StatsThe "official" recommendations by Carrie Ballinger can be found here: Statistics collection recommendations – Teradata 14.0 Teradata 13.10 Statistics Collection Recommendations &n...
279427 May 2014 @ 01:06 PDTDatabaseReplyoptimization of queryPosting the same question repeatedly will not provide you with any additional answer than the one posted by Raja: http://forums.teradata.com/forum/general/teradata-pivot-function-equivalence-of-or...
279327 May 2014 @ 09:28 PDTDatabaseReplyStatistics Collection ReplicationYou can run TSET on command line, check the "Teradata System Emulation Tool User Guide" manual, "Chapter 4: Command-Line Interface"
279227 May 2014 @ 04:10 PDTDatabaseReplyStored ProcedureThe CALL step is always the longest because it's the overall runtime, it sums up the runtimes of each SQL within the SP. In QryLogV you get the CALL step with a RequestNum x, the SQLs within t...
279127 May 2014 @ 02:51 PDTDatabaseReplyoptimization of queryIf your query runs "absolutely fine" without MIN then why do you want to use it? :-) Without additional details nobody will be able to help.  
279027 May 2014 @ 02:29 PDTDatabaseReplyExpanding the column sizeIs there any relation to the existing topic? No? So why don't you open a new topic? where col like '%a%_%a%_%a%'
278927 May 2014 @ 02:24 PDTDatabaseReplyRights held by DB/UserPress F12 to get the SQL History window where you see all SQL submitted by TD Administrator.
278825 May 2014 @ 10:23 PDTDatabaseReplyAvoid all rows scan on partial varchar accessIs there any rule how to extract info from that column like "check the first 3 chars"? Some kind of intelligent key compsed from subparts? Them splitting the key into those subparts and a...
278725 May 2014 @ 10:11 PDTDatabaseReplyHow display list of foreign key constraints, including WITH [NO] CHECK OPTION setting?There's no way to get that info from the dbc tables/views, this is only stored in the table header :-(
278625 May 2014 @ 10:08 PDTGeneralReplyInstallation of Teradata on LinuxHi Achyuth, yes, Teradata only runs on SLES. For pricing there's  http://www.teradata.com/brochures/teradata-workload-specific-platform-family/
278525 May 2014 @ 06:44 PDTDatabaseReplyWhy is this statistic recommended?Hi Joe, this might be due to better row estimations, is it a 1:many or many:many join. Based on you narration you're not joining on the UPI, which is known to optimizer.
278425 May 2014 @ 06:42 PDTConnectivityReplyteradata and password expired!When you simply look at the "ODBC Driver for Teradata User Guide" you'll find an option for SQLDriverConnect and SQLBrowseConnect() named "PWD2": "If PWD2...
278325 May 2014 @ 06:33 PDTDatabaseReplyStatistics Collection ReplicationWhat's your TD release? You can extract stats from prod and re-apply them to DR using the System Emulation Tool (TSET) In TD14 you can also simply catch the output of a SHOW STATS V...
278222 May 2014 @ 11:38 PDTDatabaseReplyString pattern match using regular expressionsHi Aarsh, you should clearly specifiy what you want. In your first example there's a number 15419928832 which is not within the expected range :-) So you need exactly 10 digits, are you ...
278122 May 2014 @ 01:48 PDTDatabaseReplySUBSTRING in ODBC Vs Teradata.Net modeHi Sau, SUBSTRING(COL1, 0,10) is ODBC syntax, based on a setting within the ODBC driver options it's automatically rewritten to valid Teradata SQL: SUBSTRING(COL1 FROM 1 FOR 10) -- ANSI Sta...
278022 May 2014 @ 12:13 PDTGeneralReplyMerge Statement for inserts onlyInsert/Select might perform a bit faster, but needs a spool, while Merge probably doesn't need spooling (check Explain). For an insert only Merge you usually add a dummy condition: MERGE IN...
277922 May 2014 @ 12:03 PDTDatabaseReplyString pattern match using regular expressionsHi Aarsh, WHERE CAST(REGEXP_SUBSTR(col, '[0-9]+') AS BIGINT) BETWEEN 5419928000 AND 5419928999 If you care about a leading sign you can add it to the regex: '[+-]?[0-9]+'

Pages