Search found 53125 matches

by ray.wurlod
Wed Apr 06, 2005 4:07 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Which is best tool for handling 20TB to 30TB Data ?
Replies: 10
Views: 2921

Go to Ascential's web site and do a search for "benchmark". There are some success stories (with big data volumes) that you can use to support your pro-DataStage case.
by ray.wurlod
Wed Apr 06, 2005 3:59 pm
Forum: Site/Forum
Topic: Forum time GMT +1
Replies: 10
Views: 5926

YOUR particular offset is configured in your profile. There is no mechanism for changing it automatically. (That's actually quite a difficult task to achieve in a world wide sense - for example some states in Australia do have summer time while others don't.)
by ray.wurlod
Wed Apr 06, 2005 12:00 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: clear log
Replies: 6
Views: 2563

Revert to status view in Director. What is the status of the job? The usual reason for the Clear Log facility being disabled is that, rightly or wrongly, the status of the job remains recorded as Running. If this is the case, and you are totally certain that the jobs are no longer running, then you ...
by ray.wurlod
Tue Apr 05, 2005 11:35 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: After-Job Routine
Replies: 15
Views: 4866

ExecDOS installs on UNIX servers. :P
by ray.wurlod
Tue Apr 05, 2005 10:27 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Generate variable output records per input record
Replies: 12
Views: 3240

FUNCTION GenerateOutputLine(Arg1, Arg2, Arg3, Arg4, Counter) * Generates as many tab-delimited output lines as Counter indicates EQUATE TAB To Char(9) If System(91) Then Terminator = Char(10):Char(13) ; * DOS End Else Terminator = Char(10) ; * UNIX En...
by ray.wurlod
Tue Apr 05, 2005 10:23 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Generate variable output records per input record
Replies: 12
Views: 3240

Yes, you use a Transformer, but almost certainly use a Routine to generate the single output column based on the input columns, which are arguments to the routines. If you just wanted empty rows you could use Str(Stuff:Char(10),Counter) but for anything much more complex a Routine is indicated. The ...
by ray.wurlod
Tue Apr 05, 2005 10:20 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Deletion of files of pattern DSD.RUN* from &PH& dire
Replies: 5
Views: 2039

Good catch!

Last time I heard, Mukund was in the UK, so might only need NLS to display the correct currency symbol!
by ray.wurlod
Tue Apr 05, 2005 7:30 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Generate variable output records per input record
Replies: 12
Views: 3240

That's actually quite a different problem, but solved fairly easily using an intermediate text file and generating the same number of line terminators as there are in your count column. You need to be careful to specify no formatting in the Inputs tab for the Sequential File stage; when you read fro...
by ray.wurlod
Tue Apr 05, 2005 7:26 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Executing SQL statements in DataStage
Replies: 8
Views: 3397

Your SQL is wrong. What you should be doing is SELECT COUNT(TableB.SID) FROM TableA,TableB where TableA.SID = TableB.SID (there are two table names in the FROM clause, and the counted column is exactly specified, which can be more efficient than * in certain databases). Build your SQL in wha...
by ray.wurlod
Tue Apr 05, 2005 7:17 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: After-Job Routine
Replies: 15
Views: 4866

If ExecDOS is not there, either someone has removed it (some people believe it's dangerous to permit access to the operating system), or your DataStage server is incompletely installed.
by ray.wurlod
Tue Apr 05, 2005 7:10 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Using Quality stage plug-in
Replies: 4
Views: 1263

Log in as qsadm and start the qsrtmngr manually. Detailed instructions are in the manual. This will allow you to determine whether this is your only problem. If it is, get your system admininstrator to configure the qsrtmngr to start automatically when all the other DataStage processes start.
by ray.wurlod
Tue Apr 05, 2005 7:07 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: CREATE.FILE and path
Replies: 7
Views: 2264

You can also use DDL. This has the side-effect of building the dictionary (as does Chuck's approach).

Code: Select all

CREATE TABLE tablename
   DATA pathname
   DICT pathname
(
   column_definition_1,
   column_definition_2,
   etc.
);
by ray.wurlod
Tue Apr 05, 2005 7:02 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Deletion of files of pattern DSD.RUN* from &PH& dire
Replies: 5
Views: 2039

You might investigate the cause of such a large stdout, so that you can address the cause and avoid its recurrence. :wink:
by ray.wurlod
Tue Apr 05, 2005 6:59 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: performance issue
Replies: 2
Views: 709

Provided you're not using too many routines, the memory address is cached, so that you really only incur the cost of loading the subroutine code ("snapping the link") once. Unless the machine is overloaded, the likelihood of actually finding the routine still in memory for the next row is very high....
by ray.wurlod
Tue Apr 05, 2005 6:56 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to handle transaction in BASIC ROUTINE
Replies: 2
Views: 982

If the routine is connecting to DataStage database (hashed files, UV tables) then you can use BEGIN TRANSACTION, COMMIT, ROLLBACK and END TRANSACTION as you seek to do. If you are accessing other databases via the BASIC SQL Client Interface (BCI), then you use SQLSetConnectOptions() to specify manua...