AI-Enabled Adaptive Fault Injection for Self-Regulating Software Testing in AWS Cloud Platforms

Fuente: Zenodo
Saved in:
Bibliographic Details
Main Author: Manuja Bandal
Format: Recurso digital
Published: Zenodo 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866901728961495040
author Manuja Bandal
author_facet Manuja Bandal
contents <p><span>Testing software in cloud environments, especially within AWS infrastructures, presents distinct obstacles due to dynamic resource allocation, decentralized architectures, and unpredictable execution conditions. Conventional testing methods often fail to detect cloud-specific faults such as transient errors, race conditions in autoscaling, and inconsistencies in distributed systems. This paper introduces AI-Enabled Adaptive Fault Injection (AIAFI), a novel autonomous testing framework specifically designed for AWS-based applications. AIAFI autonomously detects, injects, and modifies fault scenarios in cloud-native applications through reinforcement learning (RL) and evolutionary search mechanisms. By utilizing AWS-integrated observability tools such as CloudWatch, X-Ray, and AWS Fault Injection Simulator (FIS), our approach enhances fault detection by 65% compared to leading-edge testing methodologies. Experimental results demonstrate that AIAFI effectively optimizes test execution, minimizes downtime, and improves the resilience of AWS-powered infrastructures.</span></p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_15044761
institution Zenodo
language
publishDate 2022
publisher Zenodo
record_format zenodo
spellingShingle AI-Enabled Adaptive Fault Injection for Self-Regulating Software Testing in AWS Cloud Platforms
Manuja Bandal
Fault Injection
Cloud
AWS
Artificial Intelligence
Software Testing
<p><span>Testing software in cloud environments, especially within AWS infrastructures, presents distinct obstacles due to dynamic resource allocation, decentralized architectures, and unpredictable execution conditions. Conventional testing methods often fail to detect cloud-specific faults such as transient errors, race conditions in autoscaling, and inconsistencies in distributed systems. This paper introduces AI-Enabled Adaptive Fault Injection (AIAFI), a novel autonomous testing framework specifically designed for AWS-based applications. AIAFI autonomously detects, injects, and modifies fault scenarios in cloud-native applications through reinforcement learning (RL) and evolutionary search mechanisms. By utilizing AWS-integrated observability tools such as CloudWatch, X-Ray, and AWS Fault Injection Simulator (FIS), our approach enhances fault detection by 65% compared to leading-edge testing methodologies. Experimental results demonstrate that AIAFI effectively optimizes test execution, minimizes downtime, and improves the resilience of AWS-powered infrastructures.</span></p>
title AI-Enabled Adaptive Fault Injection for Self-Regulating Software Testing in AWS Cloud Platforms
topic Fault Injection
Cloud
AWS
Artificial Intelligence
Software Testing
url https://doi.org/10.5281/zenodo.15044761