Resilient Auto-Scaling of Microservice Architectures with Efficient Resource Management

TL;DR

This paper introduces SecureSmart HPA, an advanced auto-scaling system for microservices that enhances resilience and resource efficiency by detecting disruptions and dynamically adjusting scaling, significantly reducing resource wastage and improving performance.

Contribution

It extends prior work on Smart HPA by incorporating disruption detection, resource sharing, and dynamic adjustments for resilient, resource-efficient auto-scaling in microservice architectures.

Findings

Achieves up to 57.2% reduction in CPU overutilization.

Increases resource allocation by 51.1%.

Performs effectively across 25%, 50%, and 75% resource wastage levels.

Abstract

Horizontal Pod Auto-scalers (HPAs) are crucial for managing resource allocation in microservice architectures to handle fluctuating workloads. However, traditional HPAs fail to address resource disruptions caused by faults, cyberattacks, maintenance, and other operational challenges. These disruptions result in resource wastage, service unavailability, and HPA performance degradation. To address these challenges, we extend our prior work on Smart HPA and propose SecureSmart HPA, which offers resilient and resource-efficient auto-scaling for microservice architectures. SecureSmart HPA monitors microservice resource demands, detects disruptions, evaluates resource wastage, and dynamically adjusts scaling decisions to enhance the resilience of auto-scaling operations. Furthermore, SecureSmart HPA enables resource sharing among microservices, optimizing scaling efficiency in resource-constrained environments. Experimental evaluation at varying disruption severities, with 25%, 50%, and 75% resource wastage, demonstrates that SecureSmart HPA performs effectively across different levels of disruptions. It achieves up to a 57.2% reduction in CPU overutilization and a 51.1% increase in resource allocation compared to Smart HPA, highlighting its ability to deliver resilient and efficient auto-scaling operations in volatile and resource-constrained environments.