DSXchange

My motorcycle doesn't start. What's wrong with it?

If you had had the foresight to post the error messages you are getting (exactly) - that is, the symptoms of the problem - we might be in a better position to offer cogent advice. As things stand, that is not possible.

As you're providing the greater detail that Craig requested, please also advised how the data are partitioned on every input link to the Lookup stage.

I agree with Kim, why?

If it's odds and evens, you could use a two-node configuration and round-robin partitioning, or you could use a single-node configuration and a Column Generator and a Filter stage.

I would be interested to know what that "elsewhere" is. It's not totally correct. In particular, the numeric prefix is not necessarily four bytes; it may be as small as one or two bytes. Essentially, the "prefix" size in the record schema gives you this value. Storage in memory a...

A Copy stage with no output links serves as a sink and consumes no (or negligible) resources. Look at the score for the job.

Lookup File Sets can obviate the LUT_CreateOp process; it is a useful trick when the same reference data set is needed in multiple jobs.

This is a classic fork-join design. Search for "fork join".

The dsenv script must not put any unexpected information onto stdout, and probably does not necessarily capture telnet connections to the server.

Add environment variable APT_DECIMAL_INTERM_SCALE to your job as a parameter.

The calculation is exactly the same in server and parallel jobs. Please be more specific about (a) data types and (b) what you mean by "incorrect".

My solution does not require sorting.

Source table name and target table name (and schema in both cases) can be specified as parameters. Key column cannot. Therefore the answer to your final question is "no".

The only real "information" is what's in the manuals, on-line help and RedBook. Do you have any specific questions?

Welcome aboard.

Use ODBC. Warning: EVERY file on AS/400 is considered by ODBC to be a "table" so make very sure to include a name filter in the ODBC data source configuration.

Welcome aboard.

The easiest way is to:

• add an artificial key to the data using a Column Generator stage
  
  create a copy of the data stream using a Copy stage
  
  join the two copies using a Join stage (take care to separate column names)