Software Engineering - software architecture reconstruction

Author

software architecture reconstruction

Ashok

2006-10-30, 9:59 pm

People often do software architecture reconstruction to extract the
architecture of a software. The outcome as I observed is generally more
like source code structure, i.e, interactions among different modules
abstracted into high level, based on patterns like "file contains
function", "directory contains file" etc.

My confusion is, can I really call the outcome of SAR as the
architecture of a software.

Phlip

2006-10-30, 9:59 pm

Ashok wrote:

> People often do software architecture reconstruction to extract the
> architecture of a software. The outcome as I observed is generally more
> like source code structure, i.e, interactions among different modules
> abstracted into high level, based on patterns like "file contains
> function", "directory contains file" etc.
>
> My confusion is, can I really call the outcome of SAR as the
> architecture of a software.

No. The architecture of the software is always inside the software. Anything
else is a model.

The architecture of a building is in the building. The blueprints are just
symbols of the architecture.

Now what are you actually trying to accomplish?

--
Phlip
[url]http://www.greencheese.us/Z

Land[/url] <-- NOT a blog!!!

H. S. Lahman

2006-10-30, 9:59 pm

Responding to Ashok...

> People often do software architecture reconstruction to extract the
> architecture of a software. The outcome as I observed is generally more
> like source code structure, i.e, interactions among different modules
> abstracted into high level, based on patterns like "file contains
> function", "directory contains file" etc.

It depends upon what one is reconstructing from and how one does the
reconstruction. At one extreme, a well-formed OOD model in UML will
essentially already have a Component, Package, and/or Deployment
Diagrams that directly show the basic modularity of the application.

At the other extreme, if one reverse engineers the code itself it is
very difficult to extract the basic architecture since a great deal
design information has been lost in the final code. That's because the
code does not explain Why design decisions were made and it is difficult
to separate architectural decisions from tactical OOP decisions or
dependency management decisions.

> My confusion is, can I really call the outcome of SAR as the
> architecture of a software.

I don't think so. SAR is pretty specialized. The architecture is
represented by the AAL within which SAR represents a particular layer's
implementation.

OTOH, a lot depends on how one defines 'architecture'. B-)

*************
There is nothing wrong with me that could
not be cured by a capful of Drano.

H. S. Lahman
hsl@pathfindermda.com
Pathfinder Solutions
http://www.pathfindermda.com
blog: http://pathfinderpeople.blogs.com/hslahman
"Model-Based Translation: The Next Step in Agile Development". Email
info@pathfindermda.com for your copy.
Pathfinder is hiring:
http://www.pathfindermda.com/about_us/careers_pos3.php.
(888)OOA-PATH

Sponsored Links