DSXchange

It's usually called "masking". Investigate IBM's Optim product, which has this capability, while preserving the relationships and statistics of the data.

There is nothing automatic. You must design this capability into your job.

The maximum buffer size and maximum number of columns can be tuned in the uvodbc.config file on a per-DSN basis or globally. Search DSXchage for MAXFETCHBUFFSIZE and MAXFETCHBUFFCOLS.

Or you're trying naughtily to WRITE to a record for which you're not the lock owner.

One for each field.

Yes but, as noted, you can not use Char(any) in the value of a routine argument. This needs to have been pre-evaluated. Char(254) is pre-evaluated into the @FM system variable and Char(253) is pre-evaluated into the @VM system variable. But it's more difficult than that. You can not use any form of ...

(Yes, Mo - Megabyte and Go = Gigabyte. These are fairly commonly used abbreviations in southern Europe particularly.)

What Craig said. You should have sought advice first. Without restoring the necessary components (and we don't know which they are, because we don't know what you thought suitable for eradication) you will not be able to log in to this project.

Any time, driven by the metadata. For example, you might extract the following Char(15) string from source: "010017028048100". The record schema in, say, the output link of a Copy stage might specify this as a vector of five three-character fields. It's still the same 15 characters. For ta...

Create the job with a Date type job parameter.
When the job is submitted, provide a legal date value (in YYYY-MM-DD format).
For example

You are not correct. The Lookup stage maps the reference data set into memory before the first row is processed. No subsequent I/O is involved. Look elsewhere for your I/O bottleneck.

It's done with environment variables APT_ORA_UPSERT_COMMIT_INTERVAL and APT_ORA_UPSERT_COMMIT_ROWS.

Welcome aboard. Presumably the decision is made in your job logic. Best practice is to use two Connectors - one to handle the inserts and one to handle the updates. To avoid deadlocks, hold one or the other to a Data Set and move those records into the table in a subsequent job.

Now show the expression used to generate the value for the job parameter jprFormatMasks. I presume this is done in a sequence.

Also show us the entry from the job log showing the value passed to the job via that parameter.