DSXchange

been long to remember the syntax but the code should look like above

1. replace all non numeric characters to space (you can do double convert), 2. do a default trim to remove leading, trailing, repeating spaces. 3. use field function to get the first set of digits with delimiter as space. 4. take its index and add the length of first set of digits and you have the l...

you should have a surrogate key for updates in such cases where you can replace the nulls with some value you are not expecting in the data such as 0, -9999 or ~, and then do a lookup and send records to update based on surrogate key you got from there.

With such "basic transformation" you could optimize it yourself. Few years back that option wasn't there in datastage at all.

A batch ID from what I have seen is an integer to track the records being inserted/updated during Job run. Mostly for audit purposes. And that needs to be generated before each job execution, a separate piece of code.

But only OP knows what he really wants.

Seems system selected block size is taken as 1000 hence 1 to 400 and 1001 to 1401 has been generated. You can change the block size to 1 to see it being generated in a compacted manner, However it has performance implications. Surrogate key is just a unique number and should not have such requiremen...

Do you get the same warnings when executed from director, but still it completes?

Talk to your DBA to get the listener started or to get the TNS entry corrected.

Since you are on LPARs, please run below commands when you experience slowness (please use -a if -A doesn't work). sar -u 5 5 sar -r 5 5 sar -A 1 1 First 2 commands will give you CPU and memory stats 5 times every 5 seconds. In third commands output I am more interested in number of faults. I would ...

It does not generate different checksum for the same value if the data type is identical as well including unicode.

If the checksum value is different then it seems that the values are different or the data type is not identical or any padding is present while calculating the checksum. Also before join, you need to make sure that hash column has same metadata to get correct joins. Since you are not getting the co...

you can try setting APT_NO_JOB_MON = 'True' just to check if its the job monitor which is causing the problem.

Above enviornment variable changes will disable the monitoring and the statistics will not be visible in director.

10 runs isn't bad, and should not be pushing the logs to GBs however You may need to check if these auto purge settings are actually executing. However I would check if the shared memory parameters and ulimit values are at recommended level. IS DS installed on VMs/LPARs or Physical servers? did you ...

Did you check if you have any after job subroutine in that job?

Reg 11.1 support, you can always use v10 client to connect to 11.1 server if you must. Things we have been doing since long.