DSXchange

How many CPUs do you think would be required to process a query such as SELECT MAX(keyvalue) FROM table;

The answer to your overall question clearly is "it depends".

Transaction isolation levels are part of the ODBC standards. There is a good discussion in the DataStage BASIC manual. Read that, particularly the part about what potential data anomalies are avoided by each. Then, if you still have questions, post here again.

10000 rows is not large for a hashed file, particularly if you've only loaded the columns you actually need from the DB2 table.

UV stage can not connect to DB2 but ODBC can, and still deliver multi-row result sets from lookup. This is addressed in your other post.

Your support provider should be able to obtain this quickly (electronically) from IBM.

Hashed files are not necessarily pre-sized. This is a choice that the developer can make. By default table space management in DataStage dynamic hashed files is automatic.

Get the SQL data types in your job design to match those in the DB2 table (that is, Char of the appropriate size).

Simple is often the most effective. But, with this design, how do you handle the duplicate key? Or are you generating surrogate (artificial) keys into the target table?

"Serks", actually - Kim is in Texas.

Not enough information. DataStage server jobs typically send rows; the number of rows in each "batch" is set by "array size" or some similarly-named property. If what you claim is true there is some problem with your configuration; you need to involve your support provider. There is no reason to los...

One obvious option is to upgrade to 7.5, where you get all of the goodies whose absence you bemoan. What precisely do you mean by "abort the mail sequence"? What's wrong with the method you've chosen (using a trigger to detect the failure)? There's no difficulty (using such an approach) to have a wa...

I understand completely what you are trying to do, except for the specific circumstances (such as whether there is a known maximum number of columns that will be generated). The technique is called "vertical pivot". How you accomplish it depends on circumstances such as that mentioned above. There i...

I believe you'll find that no decimal placeholder character is stored at all; it is applied when extracting decimal data from the database. The fact that it is a comma suggests that you are in a European locale. If you are running DataStage with NLS enabled, you can set your locale differently. The ...

That's for non-obvious reasons. You need a fresh attachment. handle1 = DSAttachJob("job", DSJ.ERRFATAL) Status = DSGetJobInfo(handle1, DSJ.JOBSTATUS) If Status = DSJS.RUNFAILED Or Status = DSJS.CRASHED Or Status = DSJS.STOPPED Then ErrCode = DSRunJob(handle1, DSJ.RUNRES...

DataStage can do it, but you have to generate a table definition that contains data types that are legal for SQL Server. That part is not automatic. I'm not aware of any tools out there that can do it automatically.