Need expertise opinion on ADF custom BackEnd implementation

Question

Hi all,
Jdeveloepr version: Studio Edition Version 11.1.1.5.0.
I am working on an ADF application and trying to go with the following implementations which are currently working fine while unit testing (me alone- single user) . I would like to have a expert opinion as whether this would be a scalable solution for an application which will support multiple users simultaneously . I am very new to ADF framework , so please advice
-The page are mostly inside bounded taskflows, which are nested and having backing beans are different scope. The backing bean is talking to a service layer (java classes ) to call the Webservice layer few database tables and then integrate and process the data to be shown on the front end.
(we have decided to opt out of drag and drop technique in ADF, due to some UI design reasons and large Webservice calls, though it is possible as mentioned by some experts)
The questions are based on the Database interaction, we have created an appmodule which has several VOs abased on entity objects and accessing them programmatically from the service layer like
1. Get the appmoduleDatacontrol via the id name in Databindings.cpx and get corresponding VO implementation classes to invoke custom methods written inside. How good would be the connection pooling and state management of the application modue in production environments.
2. for insertion - we call createROW on VO implementation and call an commit on the row implementation classes after adding detials to the new row.
3. for select op- we create filter criteria on the view and apply them programmatically to get the rows.
4. for update- get the row using the filters VO and manipulate the row using RowImpl classes and do a commit on then. (a commit on VO or commit on RowImpl, which would be better?)
Any pointers to tutorials/Blogs related optimizing and testing this implemtnation to this would be really helpful.
Thanks,
JSWL

Answer 1

Hi all,
Jdeveloepr version: Studio Edition Version 11.1.1.5.0.
I am working on an ADF application and trying to go with the following implementations which are currently working fine while unit testing (me alone- single user) . I would like to have a expert opinion as whether this would be a scalable solution for an application which will support multiple users simultaneously . I am very new to ADF framework , so please advice
-The page are mostly inside bounded taskflows, which are nested and having backing beans are different scope. The backing bean is talking to a service layer (java classes ) to call the Webservice layer few database tables and then integrate and process the data to be shown on the front end.
(we have decided to opt out of drag and drop technique in ADF, due to some UI design reasons and large Webservice calls, though it is possible as mentioned by some experts)
The questions are based on the Database interaction, we have created an appmodule which has several VOs abased on entity objects and accessing them programmatically from the service layer like
1. Get the appmoduleDatacontrol via the id name in Databindings.cpx and get corresponding VO implementation classes to invoke custom methods written inside. How good would be the connection pooling and state management of the application modue in production environments.
2. for insertion - we call createROW on VO implementation and call an commit on the row implementation classes after adding detials to the new row.
3. for select op- we create filter criteria on the view and apply them programmatically to get the rows.
4. for update- get the row using the filters VO and manipulate the row using RowImpl classes and do a commit on then. (a commit on VO or commit on RowImpl, which would be better?)
Any pointers to tutorials/Blogs related optimizing and testing this implemtnation to this would be really helpful.
Thanks,
JSWL