DSXchange

Premium membership is only a few cents per day and serves entirely to help fund the bandwidth costs of DSXchange. I would prefer that nobody undermines this mechanism.

ramesh_c wrote:I have a doubt why ODBC Enterprise stage is not used in the parallel jobs. Instead we need to use the Oracle Enterprise Stage.

Who said? Ask them about pros and cons, and their reason for the decision.

mystuff wrote:Even the backup didnt work. what shall we do now?

Don't "they" take system backups? If not, ur skrood (as "they" say). Time to re-install.

Use Hashed File Calculator. Choose the 64-bit option. Use the command that it generates.

Approximately 19 million TB. subject to operating system limits on the size of a file. There is no record limit, except that the records have to fit within this size limit.

That's not the point, really. What other action should you properly take BEFORE re-running the job? What if some rows, but not all, had been committed to the target database? Do you want to pick up where you left off (in which case your job design must keep a record of how far it got) or do you want...

No, though I believe Red Hat ASE 3.0 also works.

Adequate, but not rigorous.

Use SQL set operator DIFFERENCE when extracting.

Yes.

Generate the score, inspect the score, and find out. Or just look at the link marker on the Sequential File stage - that alone will answer your question.

You will also need a Lookup stage. Use the Transformer stage or Filter stage to determine whether aaa is null. This stage has two outputs. The output where aaa is not null feeds a Lookup stage. The output from the Lookup stage and the "null" output from the Transformer stage are brought back togethe...

(1) Yes, depending upon your version.
(2) No.

... or, slightly more efficiently,

Compare compares the entire record and can report direction of difference (> or <) rather than just "different". Difference and Change Capture perform the same comparison, but present their output in different format; Change Capture uses subrecords.