Article: Microservices extraction through set of business processes variants Journal: International Journal of Business Process Integration and Management (IJBPIM) 2024 Vol.11 No.3 pp.186 - 198 Abstract: Managing multiple variations and versions of the same business process is a common practice in many companies. Indeed, the executions of these processes differ from each other according to human and contextual factors or also according to deliberate managerial decisions. These multiple variants describe the different paths that the activities can take according to these mentioned variations, which offer better flexibility, efficiency and adaptability for any organisation. However, these variants constitute a monolithic system with components that are strongly coupled. The latter can become more complex as it develops, with intertwined processes and tight dependencies between features. Thus all the variants of BP will be bulky and difficult to manage. Additionally, changes or updates in a monolithic system may require significant changes across BP, which may lead to regression risks or maintenance challenges. In this context, we aimed to propose an approach to migrate our monolithic system to a system with components that are loosely coupled, strongly cohesive and fine-grained. It is a multi-model approach representing a control and data dependency model and which takes as system input a set of process variants. We used three different clustering algorithms to generate our candidate micro-services. Inderscience Publishers - linking academia, business and industry through research

Title: Microservices extraction through set of business processes variants

Authors: Malak Saidi; Anis Tissaoui; Sami Faiz

Addresses: National School for Computer Science, Manouba, Tunis, Tunisia ' Faculty of Law, Economics and Management, Jendouba, Tunis, Tunisia ' Higher Institute of Multimedia Arts, Manouba, Tunis, Tunisia

Abstract: Managing multiple variations and versions of the same business process is a common practice in many companies. Indeed, the executions of these processes differ from each other according to human and contextual factors or also according to deliberate managerial decisions. These multiple variants describe the different paths that the activities can take according to these mentioned variations, which offer better flexibility, efficiency and adaptability for any organisation. However, these variants constitute a monolithic system with components that are strongly coupled. The latter can become more complex as it develops, with intertwined processes and tight dependencies between features. Thus all the variants of BP will be bulky and difficult to manage. Additionally, changes or updates in a monolithic system may require significant changes across BP, which may lead to regression risks or maintenance challenges. In this context, we aimed to propose an approach to migrate our monolithic system to a system with components that are loosely coupled, strongly cohesive and fine-grained. It is a multi-model approach representing a control and data dependency model and which takes as system input a set of process variants. We used three different clustering algorithms to generate our candidate micro-services.

Keywords: multiple variations; monolithic system; control and data dependency; micro-services.

DOI: 10.1504/IJBPIM.2024.140159

International Journal of Business Process Integration and Management, 2024 Vol.11 No.3, pp.186 - 198

Received: 01 Jul 2023
Accepted: 26 Mar 2024
Published online: 25 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Microservices extraction through set of business processes variants

Keep up-to-date