Digital Twin based Fault Analysis in Hybrid-cloud Applications

Abstract

Hybrid clouds bring together the advantages of the on-premises, private, and public cloud services. Agility, privacy, and regulatory compliance are some of the advantages of hybrid clouds. However, the key characteristics of complex hybrid cloud application such as heterogeneity, reliability, interoperability, scalability, and dynamic nature also makes old static approaches brittle. A digital twin can capture real-time (meta) data of a hybrid cloud application and supply valuable insights for efficient management of the cloud services and analyzing faults. In this paper, we propose a digital twin assisted approach to analyze the faults, interoperability, and reliability issues of a hybrid cloud-based application. We also present some metamorphic relations to analyze the reliability of our twin. Insights from the digital twin can assist developers to take proactive measures before a fault occur. The digital twin can also help detect faults and identify the cause and remediation steps through data insights. The key advantage of using digital twin is that it helps detect issues in complex hybrid cloud applications in a holistic way.CCS CONCEPTS• Software and its engineering; • Computer systems organization ? Embedded and cyber-physical systems;