on 01-20-2015 7:55 AM
Hi there,
And recently we have got new PO 7.4 system , to be frank I dont have much idea on this new PO.
and the first scenario is the source input CDF file to ECC (Idoc) via PO.
we need to maintain some table (not sure where) i.e. during mapping runtime for a combination of source field (say 4 field values) from table we need to get combinaiton of target fields values and populate to the Idoc fields.
since we use single stack here my query is
1, where should we create the tables ? ECC or DB (DB2) or ?
2, which will be the best practise wrt performance ?
3, what will be best approach here?
pls suggest.
Hi,
Another one suggestion required can we validate different values from different value mapping tables is that possible?
or this can be achived only via lookup or how?
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Thanks, the entreis is in hundreds only. so we can go with value mapping or fixed value tables.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello,
Since you are using idoc so i won't suggest you to go with any type of lookup.
Simply manitain values in some custom table in ECC and then populate idoc fields using "user exits" in idoc. Much better solution in terms of performance.
Thanks
Amit Srivastava
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
thanks for your suggestions.
these are gonna be fixed values in the tables might be we will add some more entries going forward but will not change dynamically.
how about storing this table at DB level here we use DB2, i hope we can use JDBC call for this?
Is this better then ECC call or how?
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
maintaining another table at DB2 and using JDBC lookup will be additional load. You mentioned that values are not too much so don't use fixed values else you will have to do transports every time for new addition. Go for value mapping and you can easily maintain value directly in production environment,
Thanks
Anand
Hi,
If you don't have ever changing and dynamic values you can also use value mapping. Besides maintaining t table is never considered good practice in PI as its a middleware and not repository. so go for ECC considering data management.
Regards
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello Setharaman ,
1. Best choice will be at ECC as it will be easy to maintain.
2. For improving performance create a java hasmap in PI mapping and do a single lookup to ECC and store the values in the hashmap. Further , later in the mapping you can do a local lookup to the hashmap for fetching the values. Search on SDN to find out more on this.
3. Further if you are not expecting any regular changes in that table in future then you can also create a fixed value table in PI mapping itself and fetch data from that , but for that you need to have your table data in advance and that should not chnage frequently.
Thanks
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
87 | |
10 | |
10 | |
10 | |
7 | |
6 | |
6 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.