Trimar is/was a set of tools.
One of them was used to unload ims data to sequential files.
It's a utility that needs a sysin dataset to know what to unload. (segment
name, ...)
If you don't already have this tool onsite, you should find another
solution.
An IMS DBA would be the best person to ask what to use. (someone working at
your site.)
Does trimar fast unload output all the segments to a file
sequentially?
I mean, can I write a C program to read IMS segments(using DBD) one by
one from the outputted file? Is it that simple?
My IMS related knowledge dates...
The way your process the sequential file depends on how you unload the data.
GN(get next), GNP(get next within parent), ...
The file may look like
s001
s002
s003
or
s001, s100, s300
s002, s100, s200
s003, s400
Anyway, you may not be allowed to run any tool you wish on production.
There are probably some IMS utilities that i don't know of or remember that
would
unload. For simplicity and probably because i don't master IMS, i'd go with
plain
cobol with psb+ssa to unload...and reformat to pic x some fields if needed.
(easier to make the data sqlldr compatible, what ever that implies)
MSDB's are main storage databases and are cyclic in nature if memory serve
me good.
Moving the data is one thing but keeping the same functionnality under a
different dbms is another thing.
PM
As far as i can see, there is nowhere the word DB2 in this thread.
This is my last post in this thread.
You should move to a more appropriate newsgroup.
Maybe bit.listserv.ibm-main