Implementation of SCA specification

Davor Čretnik (2013) Implementation of SCA specification. EngD thesis.

Preview

PDF Download (785Kb)

Abstract

The thesis is covering an area of solutions for medium to large (enterprise) products or systems, which are supposed to tie in many different services into one homogenous entity. These kinds of systems are problematic in sense of updating and replacing specific parts of them. This is where an implementation of SCA specifiacion comes to provide modularity and independence of a product on a large scale. The otherwise heterogen systems can be presented as a set of independent components. Each of the components, which build up the whole, is easily replacable. This means that there is no need to recompile the whole system with a compiler. Main part of the SCA implementation consists of Sca container. It's operation is explained in a simple manner, as are it's scenarios of use. In order for Sca comtainer to function properly, service factory and service registry are introduced. Their means of operation is also explained. This thesis also consists of explanations of methods, mechanisms and different means of configuring the system. Everyday programming problems such as multithreading, multiplicity and wiring of different implementations of different C++ classes into a functional system, named composite, are also explained. A composite is being run in an assigned context, which is defined by common services that are being used by all components. All elements of a composite, which are presented in XML form, are explained as well as the properties of it. In the last part of the thesis, there is a demonstration of a use case in a testing environment, explaining how the developed system behaves and performs in a real life scenario. Elements of configuration are explained step by step along with a presentation of how the system structure looks like on the filesystem.

Actions (login required)