DSXchange

In an SMP configuration one copy exists in shared memory.

In an MPP or grid configuration the rows are moved to every node.

In a word: Search

Lots of people have bumped into this code in the past.

It would certainly be easier to move the files. You can leave them on the Windows machine, but you would need to make the folder in which they live visible to the HP-UX machine, for example using old technology such as Samba, or newer technology such as LDAP. Just out of curiosity, where in this mix...

There is no limit on the number of stage variables that you can use. What do you mean by "performance"? If you have to check null you have to incur that cost, whether in a stage variable or somewhere else. The overheads of using a stage variable are nil, since to do it anywhere else (for example in ...

Welcome aboard. I presume "call " is something within your scheduler. How long are you waiting? How long do you have to wait if resetting manually using Director? Are you attempting to reset a job that is already in a runnable stage? Instead of using -wait in the reset run, try using -jobstatus , an...

1. Pipeline parallelism yes, partition parallelism no.

2. The size of each "chunk", as you put it, is tunable. Read about buffer sizes and transport blocks in the manuals.

3. Not applicable. Don't tune your buffers or transport blocks to require more memory than you have.

1. You can do it all from flat files. 2. Use the ODBC driver for text files. This is fairly easy to configure - for a given DSN you point it at the directory that contains the file and optionally specify a file name suffix (for example "*.txt" or "*.csv") to filter the files that can be processed.

why lookup reference link requires entire partition method and the master link ,hash partitioning method? They don't. These are merely the defaults, which are guaranteed to give correct results. Another possibility is that the lookup key is a single integer. In that case the stream input could use ...

That would depend entirely on what the routine does, so there's no ready answer to what is at best a hypothetical question.

Having said that, it does not matter that there are "gaps" in sequences of surrogate key values. Surrogate keys provide uniqueness and nothing else - by definition.

Have you thoroughly checked that the invocation IDs are absolutely unique everywhere, including in the controlled jobs?

Yes, you are missing a specification of what to do when the data include a NULL.

Most databases have a technique in the query language for performing this kind of recursive query. That is the easiest way to do it. If you insist on using DataStage, you are up for a lookup for each possible level of separation (the number of manages between the lowest-level employee and the CEO).

An Aggregator stage or a Transformer stage can give you the maximum from each partition.