Serverless and Multi-Cloud Lakehouse Architectures: A Unified Framework for Scalable Analytics and Intelligent Governance

Cloud-native lakehouse architectures are redefining modern analytics through the convergence of serverless computing, container orchestration, and multi-cloud strategies. Traditional data warehouses and lakes struggle with agility, cost, and governance, whereas multi-cloud lakehouses deliver elastic scalability, vendor neutrality, and intelligent workload distribution. This study presents a unified framework that leverages Terraform-driven Infrastructure as Code (IaC), Container Orchestration Framework orchestration, and serverless processing to automate provisioning and optimize cross-cloud analytics workflows. The proposed model integrates event-driven and AI-assisted optimization for cost-aware query execution across AWS, Azure, and Google Cloud. Experimental validation shows significant improvements in scalability, fault tolerance, and operational efficiency while maintaining governance and compliance across federated environments. The findings demonstrate that serverless multi-cloud lakehouses not only mitigate vendor lock-in but also provide a foundation for adaptive, intelligent, and policy-driven enterprise analytics platforms.