Hi all,
I have to deal with metadata on my processes. Most of them upper 20 parameters. My engine rule Will need to Fetch the specific parameter and its value coming from interfaces, data bases or integrations.
So I wonder what It is best way to deal with them for performance:
1) create a CDT with upper 20 fields (and It could be defined more than 50 fields)
2) create a CDT with 2 fields: key and value and work as a list.
3) Define a Map as any type objects.
Thank you.
Discussion posts and replies are publicly visible
Can you give us a bit of a background on the scenario?
In general, I recommend to go with Appian Records. But only if you want to persist that data in some database.
For pure transport, go with a CDT. And the number of fields would not matter in that case.
And use maps in case you need the ultimate flexibility.
Thank a lot Stefan!
The scenario is to gather metadata info define for a business Process, for instance number of participantes, incomes, etc) as input of a rule engine to traje a decision.
Is there any diference between a CDT with lot of fields (more than 50 sometimrs) and a CdT with key AND value parameters defined as a list. This cdts Will nota persist un database.
Best regards
A CDT is much easier to work with and I do not see any difference in performance. When you store that data in process variables, make sure to keep them short and be aggressive in terms of data management.
What volume of data are you planning for?
The volume is quite variable. Depend on the business process. Im just designing as a general solution for any business process.
Thank you
"Any business process" ??? You have high goals. BTW, this already exists. It is called Appian!
Jokes aside, I do not recommend to implement another level of abstraction on top of Appian. It will not make it better.