Published in European Conference on Software Architecture (ECSA) 2019, Paris, France, 2019
Self-adaptation is nowadays considered as one of the possible solutions to handle the uncertainties faced by software at run-time. This is especially true in the case of IoT systems. These uncertainties can, in turn, affect the system QoS (Quality Of Service). In this tool demo, we present a machine learning driven proactive decision-making tool named ArchLearner, for aiding architectural adaptation. The tool enables the given IoT system to i) automatically identify the need for adaptation at an early stage; ii) perform automated decision making for generating the best adaptation strategy; iii) gather the feedback of the selected decision for continuous improvement. It also enables the architects/developers to i) visualize the adaptation process in near real-time; ii) specify the required configurations; iii) visualize the real-time QoS data.
Recommended citation: Henry Muccini, and Karthik Vaidhyanathan. 2019. "ArchLearner: leveraging machine-learning techniques for proactive architectural adaptation". In Proceedings of the 13th European Conference on Software Architecture - Volume 2 https://github.com/karthikv1392/karthikv1392.github.io/blob/master/files/ecsa2019.pdf