Abstract

Cloud-native software systems deliver scalable, flexible, and resilient services by leveraging microservices, containerization, orchestration, and dynamic infrastructure. However, inherent complexity, distributed components, and varying failure modes demand systematic resilience engineering to maintain service continuity, fault tolerance, and rapid recovery from disruptions. Resilience engineering models provide structured frameworks to anticipate, absorb, adapt to, and rapidly recover from failures in cloud-native ecosystems. These models integrate concepts from fault tolerance, distributed systems theory, chaos engineering, and adaptive control to design systems capable of withstanding component failures, network partitions, spikes in load, and operational errors. This paper examines resilience engineering models tailored for cloud-native software systems, covering theoretical foundations, architectural patterns, and practical mechanisms. A comprehensive literature review traces the evolution of resilience practices from early fault tolerance and self-healing systems to contemporary cloud-native strategies that embrace automated recovery, observability, and adaptive resource scaling. We propose a research methodology for evaluating resilience models across dimensions of failure coverage, performance overhead, detection latency, and operational complexity. The analysis highlights advantages such as improved uptime, graceful degradation, and adaptive capacity, alongside disadvantages including complexity and cost. Through results, discussion, and case synthesis, the paper elucidates best practices and outlines future research directions to enhance resilience in increasingly dynamic cloud-native environments