Authors:
Aditya Rautaray
Addresses:
Department of Cloud Solutions Security, CVS Healthcare, Ashburn, Virginia, United States of America.
Abstract:
Traffic burstiness is a significant challenge. Serverless computing provides unlimited scalability, but to maintain reliability across geographically dispersed microservices, it requires on-demand scaling. self-repairing serverless applications that require high availability in global deployments. The paper presents a Serverless Resilience Fabric, a specialized infrastructure layer responsible for failover. and dynamic load balancing without manual effort. An experimental confirmation uses Cloud Sim-Serverless to simulate a distributed e-commerce backend across a range of scenarios using a comprehensive simulation framework. This paper discusses. This report serves as a blueprint. From an interpretive standpoint, experimental results demonstrate a non-negligible decrease in tail latency and a significant increase in system availability when experiencing a flash crowd compared with traditional hash-based load balancing. The proposed Fabric has an L7-layer adaptive circuit-breaking mechanism that can monitor to a certain extent. regional health scores to smartly reroute traffic out of failing zones before they fail. behavior of fast architectures under different stress loads. The trace contains 416 distinct bursty traffic samples generated from synthetic trace logs that mimic the statistics of metrics such as. The study focuses on the problem of partial failures in a multi-region configuration. such as cold start latency, execution time, and inter-region network overhead.
Keywords: Serverless Computing; Resilience Engineering; Burst Traffic; Multi-Region Failover; Adaptive Circuit Breaking; Partial Failures; Multi-Region Configuration; Execution Time.
Received on: 26/03/2025, Revised on: 21/07/2025, Accepted on: 12/09/2025, Published on: 05/05/2026
DOI: 10.64091/ATICS.2026.000256
AVE Trends in Intelligent Computing Systems, 2026 Vol. 3 No. 2 , Pages: 68-76