Flow Oriented Modeling - Creating a Data Flow Model

Flow models focus on the flow of data objects as they are transformed by processing functions. Derived from structured analysis,flow models use the data flow diagram, a modeling notation that depicts how input is transformed into output as data objects move through the system. Each software function that transforms data is described by a process specification or narrative. In addition to data flow, this modeling element also depicts control flow.

Data flow oriented modeling is the most widely used analysis notation. Flow oriented modeling focuses on structured analysis and design, follows a top to down methodology and uses a graphical technique depicting information flows and the transformations that are applied as data moves from input to output.

The modeling tools that are used to build a data flow oriented model include context diagrams, data flow diagrams, entity relationship diagram, control flow diagram, state transition diagram, data dictionary, process specification and control specification.

Steps to create a data flow model
- Diagram 0: develop a context diagram.
- Decompose the Process into high level processes.
- In parallel to this, develop data flow diagrams, entity relationship diagrams and state transition diagrams.
- Define data stores which includes normalization.
- Develop data dictionary.
- Finalize data flow diagrams, entity relationship diagram and state transition diagrams.
- Develop process specifications which includes PDL, decision tables or trees.
- Perform transformational analysis which includes developing structure charts.
Information flow continuity must be maintained as each data flow diagram level is refined. This means that input and output at one level must be the same as input and output at a refined level.

How do we map data flow into a software architecture?

In architectural design, the mapping method uses data flow characteristics to derive a commonly used architectural style. A data flow diagram is mapped into program structure using one of the two mapping approaches:
- Transform Mapping
- Transaction Mapping
Structured design is often characterized as a data flow oriented design method because it provides a convenient transition from a data flow diagram. This type of information flow is the driver for the mapping approach.
Transform Flow: The overall flow of data occurs in a sequential manner and follows one, or only a few straight line paths. When a segment of a data flow diagram exhibits these characteristics, transform flow is present.

Transaction Flow: Information flow which is characterized by single data item called a transaction that triggers other data flow along one of many paths. When a data flow diagram takes this kind of form then transaction flow is present.

TRANSFORM MAPPING

It is a set of design steps that allows a data flow diagram with transform flow characteristics to be mapped into a specific architectural style.
- Review the fundamental system model.
- Review and refine data flow diagrams for the software.
- Determine whether the data flow diagram has transform or transaction flow characteristics.
- Isolate the transform center by specifying incoming and outgoing flow boundaries.
- Perform first level factoring.
- Perform second level factoring.
- Refine the first iteration architecture using design heuristics for improved software quality.

TRANSACTION MAPPING

In transaction mapping, information flow along two of the three action paths accommodates additional incoming flow. Each action path flows into a single transform, display messages and status. The design steps for transaction mapping is similar with a major difference lies in mapping of data flow diagram to software structure.
- Review the fundamental system model.
- Review and refine data flow diagrams for the software.
- Determine whether the data flow diagram has transform or transaction flow characteristics.
- Identify the transaction center and the flow characteristics along each of the action paths.
- Map the data flow diagram in a program structure amenable to transaction processing.
- Factor and refine the transaction structure and the structure of each action path.
- Refine the first iteration architecture using design heuristics for improved software quality.

Once an architecture has been derived, it is elaborated and then analyzed against quality criteria.