DSXchange

The Columns used in a DataStage job are not dynamic. Your choices are therefore to continue to use 25 separate jobs (since you've already created them) or to use generic (VarChar) columns - but even then you still must have the correct number of columns (you could change the shorter queries to selec...

Should this be a decimal number (3.4 months) or an integer number of completed months (3 months)?

It's easiest done with a routine. I'm sure you can construct the algorithm. Do so (on paper), then try translating into DataStage BASIC.

I suspect that it would be handled, maybe with warnings. PX starts processes, and does not care about which CPU any particular process is running on. However, this answer assumes that dynamically allocating LPARs copes gracefully with already-running processes if the number of CPUs is reduced; no re...

What has changed? "Nothing" is not the correct answer.

A hashed file (note hashed, not hash) is a database table. Primary key values can not be null. Somehow your design seems to be trying to write a null key to a hashed file.

Why do you think you need hashed files between the rest of your design and the DB2 stages?

There's also doing the arithmetic. A hashed file is made up of a number of pages (groups) the size of which you decide (either 2KB or 4KB). You have a certain volume of data that you need to store; you also allow, say, 20% headroom. So, to store N bytes of data, you need (N + 20%) / (page_size) page...

If both sources are Teradata, can you perform the join within Teradata? If not, can you reverse the logic, so that your lookup table only contains 200000 rows? Otherwise, do consider a server job, which can perform lookups directly against the (large) Teradata table. With only 200000 records to proc...

DataStage client software is (since 7.0) written using .NET Framework 1.1 (therefore the .NET framework must be installed in order for it to be run).

The log (hashed) file and the status (hashed) file are totally separate entities. Clearing the status file has no effect on the log file. Purging the log file has no effect on the status file.

You will notice that RT_STATUS2271 was cleared - not RT_LOG2271.

Parallel Maps Installation Button is disabled?

Are you logged in with Administrator access?

Welcome aboard. :D That is an entirely unfair question. The only honest answer is that there is, in existence or in potentio, an infinite number of test plans. (You did not restrict your universe of discourse. I shall, and restrict myself to ETL processing.) Test plans have granularity, from "seems ...

Read the file as a single column. Use Field() (if delimited) or Substrings() (if fixed width) to isolate the selector value. Four separate outputs as before, each with the single column that you read, and constrained by the selector value. On all five Sequential File stages, specify either no delimi...

"/" for UNIX!

Wouldn't an ExecuteCommand activity executing head -1 filepath be easier? You could just pick up its return value.

Intelligent. This is not the first attempt they've made at Intelligent Assistants. The first attempt was only (iirc) for 7.1 on Windows, and were wizards that guided you through job construction (source table? target table? voila! a job that moves data). The latest lot seem to be XML exports of para...

Assuming that there aren't too many lines in the file, you could read the entire file into an array, or a delimited string, using a Routine activity, then use something like a Field() function to extract the appropriate line number. Make use of the fact that DataStage BASIC translates line terminato...