Modelling Deployment Using Feature Descriptions and State Models for Component-Based Software Product Families

Abstract

Abstract. Products within a product family are composed of different component configurations where components have different variable features and a large amount of dependency relationships with each other. The deployment of such products can be error prone and highly complex if the dependencies between components and the possible features a component can supply are not managed explicitly. This paper presents a method that uses the knowledge available about components to ensure correct, complete, and consistent deployment of configurations of interrelated components. The method provided allows the user to perform analysis on the deployment before the deployment is performed, thus allowing error prevention before making any changes to the system. The method and model are discussed and presented to provide an alternative to current component deployment techniques.

Evaluating the release, delivery, and deployment processes of eight large product software vendors applying the customer configuration update model

“…The most commonly reported causes for deployment problems are faulty configurations, incompatible updates, and customisations. By implementing the key practices stated for the deployment process, these problems can partly be avoided [13]. If anything can be learnt from this research, it is that software vendors must integrate their CRM, PDM, and SCM [4] systems to automate the processes related to CCU.…”

Improving the customer configuration update process by explicitly managing software knowledge

“…A number of knowledge management and distribution tools are currently under development for the research project. One such tool is a tool that enables modelling of the deployment process for feature oriented component based software products [2] before the deployment takes place. This enables the user or system manager to evaluate the effects and restrictions before deploying a software product by asking this proof of concept tool "what-if" questions.…”

“…The platform serves the research as a resource for case studies, but also as a forum where Academia meets the industry to exchange knowledge. Another resource for case studies and work in the area of product contexts, is the Netherware project 2 where students (towards the end of their studies) are encouraged to develop their individual product and start a software company. Netherware contributes to the research by providing different products in different contexts that can be evaluated freely and openly.…”

“…Customer configuration updating is defined as the combination of the vendor side release process, the product or update delivery process, the customer side deployment process, and the usage and activation process. To begin with, these processes are themselves highly complex considering vendors have to deal with multiple revisions, variable features, different deployment environments and architectures, different distribution media, and dependencies on external products [2]. Also, there are no tools available that support the delivery and deployment of software product releases that are generic enough to accomplish these tasks for any…”

Deploying Component–Based Applications: Tools and Techniques

“…In the case of software deployment, the MDA approach starts with a platform-independent model of the target environment and the transformations finish with specific deployment configurations for the considered component-based applications [40]. Model-driven deployment has gained a lot of attention in recent years ( [5], [30], [40], [41], [42], [43]). In particular, OMG D&C Specification [5] follows MDA concepts to provide a unified technique for the deployment of component-based applications into distributed environments.…”