Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{204215,
        author = {Amit Kumar and Ms. Sarika Madavi},
        title = {AI-Based Resource Allocation in Cloud Computing},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {1},
        pages = {791-798},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=204215},
        abstract = {Cloud computing has transformed the digital landscape with the aid of offering scalable, on-demand, and value-effective computational services for corporations, industries, academic establishments, and governments. As cloud infrastructures hold to make bigger hastily, efficient resource allocation has come to be one of the maximum important challenges in keeping gadget performance, lowering operational fees, and ensuring most advantageous usage of computing resources. conventional aid allocation strategies regularly rely upon static rules, heuristic scheduling, and manual optimization strategies, which are more and more incapable of dealing with the dynamic, heterogeneous, and massive-scale nature of current cloud environments. In recent years, artificial Intelligence (AI) has emerged as an effective solution for smart and adaptive resource management in cloud computing systems.
This studies paper offers a comprehensive examine of AI-based useful resource allocation strategies in cloud computing and examines how machine studying, deep gaining knowledge of, reinforcement mastering, and smart optimization algorithms enhance cloud infrastructure efficiency. The examine explores the combination of AI models into cloud aid control frameworks for dynamic workload prediction, digital device placement, electricity-green scheduling, challenge migration, server consolidation, and self-sufficient selection-making. AI-driven structures can constantly analyse actual-time cloud records, identify workload styles, expect future resource demands, and optimize computational assets with minimal human intervention. Such intelligent mechanisms appreciably improve satisfactory of carrier (QoS), lessen latency, limit power consumption, and enhance usual device reliability.
The paper investigates diverse AI methodologies utilized in cloud environments, including supervised getting to know algorithms for workload forecasting, unsupervised learning for clustering and anomaly detection, deep neural networks for complex aid prediction, and reinforcement studying for adaptive and self-getting to know resource scheduling. moreover, nature-stimulated optimization processes which include Genetic Algorithms, Ant Colony Optimization, Particle Swarm Optimization, and hybrid wise fashions are discussed for fixing massive-scale cloud allocation problems. Comparative analysis reveals that AI-based totally strategies outperform conventional allocation techniques in phrases of scalability, reaction time, useful resource utilization, fault tolerance, and operational value discount.
Further, the paper highlights the developing position of AI in present day cloud paradigms consisting of facet computing, fog computing, multi-cloud structures, box orchestration, and serverless architectures. sensible orchestration platforms powered by using AI are able to balancing workloads across geographically allotted infrastructures while maintaining high availability and coffee power consumption. The research also discusses how predictive analytics and independent useful resource provisioning aid sustainable cloud computing via lowering carbon emissions and improving facts center energy efficiency.
regardless of the vast improvements, several research challenges stay unresolved. issues associated with data privacy, security vulnerabilities, model transparency, computational complexity, scalability constraints, training overhead, and real-time decision-making hold to affect the deployment of AI-primarily based cloud management structures. The paper critically examines those obstacles and discusses the need for explainable AI, lightweight studying frameworks, and relaxed federated intelligence models for future cloud ecosystems.
Subsequently, the take a look at outlines future research instructions related to self-adaptive cloud infrastructures, self-sustaining AI-pushed orchestration, federated studying-enabled cloud intelligence, virtual twin integration, and generative AI-assisted useful resource management. those rising technologies are expected to redefine the future of cloud computing by using permitting absolutely automated, smart, and power-conscious resource allocation mechanisms. standard, this paper demonstrates that AI-based aid allocation is a transformative method capable of improving the performance, scalability, reliability, and sustainability of next-generation cloud computing environments.},
        keywords = {Artificial Intelligence, Cloud Computing, useful resource Allocation, machine learning, Deep getting to know, Reinforcement mastering, assignment Scheduling, Virtualization, smart Orchestration, Multi-Cloud structures.},
        month = {June},
        }