Application symbols on system architecture diagrams
You use the Application symbol to represent an entire application and graphically show on this diagram how it is related to externals and data stores.
Within the application definition, you can specify overall information about the application, such as the process threads in the organization that it enables and the type of team effort being used to build it. To specify more details on the implementation of the application, you can create child data flow diagrams or UML diagrams, depending on the nature of the application, for example, structured or object-oriented or component-based applications.
Approach
Trust-Line Approach
The Trust Line approach is presented and described fully in the book, "Building Cooperative Processing Applications Using SAA", by John Tibbetts and Barbara Bernstein. In short, you can establish a trust line in your application, splitting your application into an "Access Region", which provides users with unfettered access to data, and a "Control Region", within which the validity and integrity of corporate data is protected. The access region is typically a user's workstation, where users can enter, replicate, manipulate, and display data at will, but this data is not the official version of the data from the enterprise perspective. For example, an Account Manager can create four different versions of the same sales order to try out different solutions to a customer's problem. The enterprise considers them trial orders until they enter the control region. In the control region, the application enforces rigid validity checks to transactions.
Within the Trust-Line Approach property, you can textually describe the Trust Line approach to the application -- where the Trust Line placement is, if data has been validated at all in the access region, and so forth.
Application Functions
This property is only used with the Process Chart process flow diagram. You can specify the Process Threads that the application performs, by clicking the Choices button, and dragging and dropping existing Process Threads into the list, or typing in new Process Threads and clicking the Add button. A Process Thread represents a major or minor business process flow through the organization. It can have one or more Process Chart diagrams reporting to it.
Application detail
Cooperative Profile
Cooperative processing design involves splitting an application into tiers that cooperate with one another. Examples are one-, two-, three-, and, in general, n-tiered applications. Within this text field, you can specify the profile for the n-tiered application.
Examples of one- and two-tiered profiles are:
▪Monolithic Central, where presentation, application logic, and data access are performed on the server, and a monitor provides the user interface on the client. An example of this profile is the typical mainframe or minicomputer design.
▪Monolithic Workstation, where presentation, application logic, and data access are all provided on the client. An example of this profile is the simplest form of computer workstation, which performs all processing itself.
▪Distributed Presentation, where presentation logic is provided on the client, and presentation, application logic, and data access are performed on the server.
▪Remote Presentation, where presentation is provided on the client, and application logic and data access are performed on the server.
▪Distributed Function, where presentation and application logic are performed on the client, and application logic and data access are performed on the server.
▪Remote Data Management, where presentation and application logic are performed on the client, and data access are performed on the server.
▪Distributed Data Management, where presentation, application logic, and data access are performed on the client, and data access is performed on the server.
An example three-tiered profile is:
▪Application-Transaction-Database, where presentation and application logic are performed on the client, application logic is performed on a transaction server, and data access is performed on a database server.
Distribution Summary
If the application is distributed, you can use this property to textually describe how the application's presentation, application logic, and data layers are distributed, such as on what nodes or servers.