#DateForumTypeThreadPost
3002 Mar 2007 @ 06:13 PSTConnectivityReplyUser defined function for currency conversions or rate lookupsYou probably could make use of a look up table .........To give a sample ..CREATE TABLE CrcyConv(TargetCrcy CHAR(2),StandardCrcy CHAR(2),End_Date DATE,ExchngRate FLOAT)PRIMARY INDEX (TargetCrcy, S...
2901 Mar 2007 @ 10:52 PSTToolsReplyrow hash locksInserts require row hash write locks.This required for purposes such as duplicate row checking, duplicate value checking(for unique) and not to mention to avoid any read transactions from occurring...
2801 Mar 2007 @ 10:44 PSTAnalyticsReplyhow to extract only mondays in 2007This query will give you all mondays in 2007SELECT CALENDAR_DATE FROM SYS_CALENDAR.CALENDAR WHERE DAY_OF_WEEK = 2 AND YEAR_OF_CALENDAR = 2007;So if you want to pick records from some other table (S...
2701 Mar 2007 @ 10:36 PSTAnalyticsReplyKilling queries in TeradataI don't think this should be happening. This looks like a case that the application is not closing the DB connections properly. Does the same thing happen if you connect via SQL assistant or so ?. ...
2601 Mar 2007 @ 12:25 PSTDatabaseReplyIs god is creating these new tables???????You can also "pray" to GOD not to create those error tables by using the "WITH CHECK OPTION" :-) It's just that God won't grant your prayer if the child table has records which violate the RI cons...
2528 Feb 2007 @ 12:01 PSTAnalyticsReply[PROBLEM] Teradata ADS Generator...From your Teradata admin can you try run this ? you may be spared from doing it all the time.MODIFY USER userid AS DEFAULT ROLE = ALL;
2426 Feb 2007 @ 11:34 PSTDatabaseReplyURGENT- CHECK CONSTRAINTYour original post mentioned that you wanted it between 1 and 7..... you may want to check the conditions on the constraint.
2326 Feb 2007 @ 11:18 PSTAnalyticsReplyneed to Compare header date in file 1 with system date in file 2You can do it via bteq, import the file contents to a table, do an SQL comparison, use bteq .if statements to do further processing etc.But don't you think it's better to do this job via your batch...
2225 Feb 2007 @ 10:20 PSTDatabaseReplyOuter Join Explain PlzOk let's assume you are doing an full outer join on employee and dept tables based on deptno.So for deptnos in employee table which does not have a record in dept table the column values (this is t...
2123 Feb 2007 @ 12:46 PSTAnalyticsReplyHow do i drop multiple tablesThere's a quick trick which DBAs normally use to generate all those drop statements.SELECT 'DROP TABLE ' || TRIM(DATABASENAME) || '.' || TRIM(TABLENAME) || ';'FROM DBC.TABLESWHERE DATABASENAME='s_m...
2021 Feb 2007 @ 11:01 PSTDatabaseReplyUrgent - Is this use of CASE statement in Teradata correctThe kind of case statement that you are trying to use works only in stored procedures (and that too not in a select query, but the normal "switch - case" like in C/C++/JAVA ).You probably need some...
1919 Feb 2007 @ 05:37 PSTDatabaseReplyfastload error:You may also need to use the INDICATORS keyword in your begin loading statement.
1819 Feb 2007 @ 12:22 PSTDatabaseReplyfastload error:use FORMAT FASTLOADin your export script.
1719 Feb 2007 @ 12:18 PSTDatabaseReplyAdding two new nodes / PPI considerationsThe PPI Orange books talks about the scenario where performance degradation can occur while redistribution.
1616 Feb 2007 @ 12:24 PSTDatabaseReplyAdding two new nodes / PPI considerationsI have heard that PPIs go through severe performance degradation while re-distribution that could be one.
1516 Feb 2007 @ 12:07 PSTDatabaseReplyPPIhttp://www.teradataforum.com/teradata_pdf/b035-1094-096a.pdf
1415 Feb 2007 @ 10:18 PSTDatabaseReplyPPIAnd to add to what Fred said, in real world scenario, don't try partitioning on a character column like that. You may be in for a lot of surprises when you try to insert some meaningful data into i...
1315 Feb 2007 @ 06:43 PSTDatabaseReplyPPI( You didn't do that exercise I told you ? did ya ? )Ok so we have an agreement that it takes the same amount of effort to decide to which AMP a row should go whether the table is PPI or non PPI.Ok...
1215 Feb 2007 @ 12:34 PSTDatabaseReplyFASTLOADI think that's the expected output. You will have to store it as 001.78 in the file and use CHAR(6) to get the results your are looking for.If you can't afford to change the input data file, then I...
1114 Feb 2007 @ 11:42 PSTDatabaseReplyNeed to convert a TimeStamp column value to Simple Date FormatBob has already given you the answer, except that I think week_of_month doesn't really get you to second Saturday because week_of_month is zero for the first partial week. The better approach is to...
1014 Feb 2007 @ 11:22 PSTDatabaseReplyPPIYou got it right to the point that it's easy to decide which AMP the record should go for 409.Ok now let's talk about the difference between PPI tables and NON PPI tables.In non PPI Tables, the tab...
914 Feb 2007 @ 08:38 PSTDatabaseReplyPPIOk let's forget about the update example, now let me put a question across to you (Ok it's my turn :-) ).I am inserting a record (empno, deptno) as (409, 33);Now tell me that if empno was UPI (un...
814 Feb 2007 @ 07:19 PSTDatabaseReplyPPILet's say that empno was actually a UPI, now given an empno, we know which AMP it will go to (hashing), but we don't know which partition it belongs, without the deptno information.So let's say you...
714 Feb 2007 @ 06:38 PSTDatabaseReplyInterval ErrorThis is sort of a funny situation, I don't pretend to know the answer, but I will explain my findings.First thing would be to do an EXPLAIN on these SQLs and you would probably have a shock to see ...
614 Feb 2007 @ 06:12 PSTDatabaseReplyPPISuppose you had an employee table with Primary index as Empno, and you wanted to partition it by deptno, you will have to make empno a NUPI though empno was unique. I believe this restriction was p...

Pages