FULL PDF
Abstract
In response to the growing complexity of multi-cloud architectures, this paper introduces a resource-aware orchestration framework designed to support adaptive service deployment across heterogeneous cloud providers. The proposed framework dynamically aligns application workloads with the most suitable cloud resources based on real-time monitoring of computational capacity, latency, and cost parameters. Leveraging machine learning for predictive scaling and policy-driven decision-making, the framework enhances performance, reduces operational cost, and minimizes service disruption. Evaluation using a simulated multi-cloud testbed demonstrates significant improvements in resource utilization and response time adaptability. This work contributes a scalable, intelligent orchestration layer suitable for evolving service requirements in dynamic multi-cloud ecosystems.
View more >>