DSXchange

The upper limit on the size of a Lookup File Set is limited only by whether the operating system supports large files. Assuming that it does not, and imposes a 2GB per file limit, a Lookup File Set can have up to 10,000 data files per partition. That is, 20,000 GB per partition.

Compared to what? Strict Unicode uses two bytes of storage per character. Other encodings, such as UTF-8, use one, two, three or four bytes of storage per character. Single-byte character sets, as the name suggests, use one byte of storage per character. (There is also a variant of Unicode that use ...

Yes it is. You may need to create/convert your hashed files with 64-bit addressing, but that's OK. Such hashed files can theoretically support 19 million TB of data (but most operating systems or file systems restrict this somewhat).

You will need special handling for "All" - probably in a separate stream in the job that you can bring back together with the main stream in a Link Collector stage or by writing to an intermediate hashed file.

I have certainly used more than 150. As Craig notes, there is no theoretical limit, but the control used in the GUI may impose a limit. If so, I imagine this would be in the thousands.

To explain: every job log is a separate database table in the Repository database for that project. There is no such thing as a project-wide log, and definitely no such thing as the job log for all projects. Further the list of jobs in a project is variable over time, so even an approach such as cre...

You're in the parallel forum OK, but you marked the job type as server.

You need an intermediate stage to effect the mapping. A Column Import stage would be good.

Welcome aboard.

Is this really a server job? Your reference to APT_PMsectionLeader, APT_StatusOK and Orchestrate suggest that it is not.

":C_1" and ":D_2" are not valid Oracle parameter markers. I expect this is the source of your problems.

They can sell you bigger hardware.

The Copy stage, if it is suitable for the required task, is infinitely to be preferred to a Transformer stage performing the same task. The same remains true of the Modify stage, but less so for Filter and Switch stages.

Convert each timestamp to a number of seconds since an arbitrary zero point, then subtract.

Whoever wrote the stage chose not to include support for the Unicode character set. This might be because of the size of the Unicode character map, or because there are more well-targeted subsets of Unicode (such as UTF-8) available.

That's why I also gave a DIR command. Many Windows sites have a UNIX emulator such as MKS Toolkit that could be used to execute the ls command.

In a job sequence.