A cybernetic case study of viable computing systems: A framework for requisite variety via algorithmic hot-swapping

Abstract

This paper presents a case study considering algorithmic hot swapping in the context of research surpassing Autonomic Computing, towards Viable Computing Systems. Cybernetic, mathematical and biological metaphors are allied to the human autonomic agent capability of the managerial cybernetics underscoring Beer's Viable System Model (VSM). A dual-perspective set theory design grammar model is employed exhibiting relationships between the systems and the recursive levels of the VSM. By incorporating the environment as part of the system, the technique promotes both portability and viability within an initially closed, yet changing, environment. Algorithmic hot swapping has been used to provide a repertoire of tailored responses to environmental change within this context. Systemic emergence and viability is thereby promoted, whilst an associated Learning Classifier System (LCS) is suggested to allow the system to develop an adaptive environmental model of appropriate, optimized responses, similarly demonstrating proof of the temporal and autonomic properties of the VCS concept.