Mitigating Cloud Disruptions: An AI-Driven Approach to Proactively Assess and Resolve Impact on Customer Workflows
Manoj Kumar Singhal, Chhaya Gunawat · 2024
Cloud customers frequently encounter disruptions due to changes implemented by cloud providers. These changes, often intended to enhance security and maintain optimal performance, can inadvertently degrade the performance of existing workflows or cause unexpected failures. This predicament raises questions about the responsibility: Is it the customer's fault for being impacted despite not altering their setup, or the cloud provider's for deploying changes that inadvertently affect users? Regardless of the source, the ultimate burden falls on the end user. To address this issue, we propose the development of an AI model designed to evaluate incoming changes and proactively assess their potential impact on cloud workflows. This model would 1) Alert Cloud users in advance, enabling them to prepare and mitigate any adverse effects. 2) In case disruptions occur, the model could identify the specific code or component causing the problem and suggest solutions. By implementing this AI-driven approach, cloud providers can continue to deploy essential updates and security patches without negatively impacting customers, while providing cloud users with actionable insights to maintain smooth operations.