OS/390 Connection

Post questions here relative to DataStage Server Edition for such areas as Server job design, DS Basic, Routines, Job Sequences, etc.

Moderators: chulett, rschirm, roy

Post Reply
raviyn
Participant
Posts: 57
Joined: Mon Dec 16, 2002 6:03 am

OS/390 Connection

Post by raviyn »

Hi ppl,
I have a doubt regarding Datastage 390.Is my logic correct.
We have got ODBC drivers(be it windows or unix flavours) for DB2 on OS/390 through which i can connect to mainframes.Does this essentially means i can do away with 390 component ..i.e i can use ODBC stage in server jobs itself for retreiving data from the mainframes.
If the above is true then when does DS390 come into the picture. What would be the possible scenarios where i would be using it...Is it something like whenever i have target as mainframes then we need to use it for better performance......
Please enlighten me on this topic.
[?][?][?][?][?]
Viswanath
Participant
Posts: 68
Joined: Tue Jul 08, 2003 10:46 pm

Post by Viswanath »

Hi Ravi,
We have used The DS390 version extensively. If you know the way DS390 version works, you will see that the COBOL code and JCL for running the COBOL code are generated by DataStage. This will be useful when your processing is totally mainframe related. In that case you do not have to create a connection between the DataStage server and the DB2 client and proceed for procesing. I hope this is clear. This is will put pressure on the processing time. Also the DS390 versiion gives you many features for the DB2 flavor is not present in the server version. Like DB2 Bulk load which is a very useful feature. The file types on mainframe also varies. This is taken care by the various file types available with the DS390 version. So, if you are doing an ETL process from the Mainframes to Mainframes, always wise to use this version.

Viswanath

Viswanath.S
kjanes
Participant
Posts: 144
Joined: Wed Nov 06, 2002 2:16 pm

Post by kjanes »

Just a note of caution. If you use DS server and connect to mainframe data, you may run into logic/processing that cannot effectively be done on the server side. As noted by viswanath, it is a good idea to code mainframe type jobs using the 390 component.

Remember, the DS server pallete was designed to process windows/aix data and structures, not mainframe data. The 390 pallete is definately tailored for mainframe ETL.

Kevin Janes
Post Reply