Reinforcement learning for data center energy efficiency optimization: A systematic literature review and research roadmap

Applied Energy, Journal Year: 2025, Volume and Issue: 389, P. 125734 - 125734

Published: March 26, 2025

Language: Английский

---

A Proposal for a Tokenized Intelligent System: A Prediction for an AI-Based Scheduling, Secured Using Blockchain

Systems, Journal Year: 2024, Volume and Issue: 12(3), P. 84 - 84

Published: March 6, 2024

Intelligent systems are being proposed every day as advances in cloud increasing. Mostly, the services offered by these dependent only on their providers, without inclusion of from other specialized third parties, or individuals. This ‘vendor lock-in’ issue and limitations related to offering tailored could be resolved allowing multiple providers individuals collaborate through intelligent task scheduling. To address such real-world systems’ provisioning executing heterogeneous services, we employed Blockchain Deep Reinforcement Learning here; first is used for token-based secured communication between latter predict appropriate scheduling; hence, guarantee quality not immediate decision but also long-term. The empirical results show a high reward achieved, meaning that it accurately selected candidates adaptably assigned tasks based job nature executors’ individual computing capabilities, with 95 s less than baseline completion time maintain Quality Service. successful collaboration parties this tokenized system while securing transactions predicting right scheduling makes promising advanced use cases.

Language: Английский

---

Application-Oriented Cloud Workload Prediction: A Survey and New Perspectives

Tsinghua Science & Technology, Journal Year: 2024, Volume and Issue: 30(1), P. 34 - 54

Published: Sept. 11, 2024

Language: Английский

---

An Optimized Beluga Whale Approach for Migration to Reduce Power and Service Level Agreement in Real-Time System

Contemporary Mathematics, Journal Year: 2024, Volume and Issue: unknown, P. 27 - 40

Published: Dec. 30, 2024

Managing power consumption in cloud data centers has become a critical challenge. Live container migration is technology supporting energy efficiency this context. To address the hurdles of management, thereby mitigating carbon emissions and minimizing service level agreement (SLA) violations, we propose an approach utilizing real-time server with Beluga Whale Optimization Algorithm (BWOA) for migration. The proposed aims to optimize while ensuring SLA compliance. BWOA, machine learning-based method, employed predict resource requirements containers migrate them hosts sufficient resources. We implemented compared its performance other algorithms terms response time. results demonstrate remarkable 30% improvement time, leading reduced violations optimized containerized centers.

Language: Английский

---

A Hybrid Machine Learning Approach to Cloud Workload Prediction Using Decision Tree for Classification and Random Forest for Regression

International Journal of Scientific Research in Computer Science Engineering and Information Technology, Journal Year: 2024, Volume and Issue: 10(6), P. 2240 - 2252

Published: Dec. 31, 2024

The dynamic nature of cloud workloads necessitates accurate predictions to optimize resource utilization, enhance performance, and ensure quality service (QoS). Consequently, numerous researchers have developed workload prediction models improve design deployment. These enable timely reliable forecasting, facilitating critical decisions such as allocation network bandwidth management. This study proposes a hybrid learning model, termed DTCRFR, which integrates Decision Tree Classification Random Forest Regression techniques predict workloads. DTCRFR model operates by initially assigning state each input data point based on historical system metrics. Subsequently, the regression refines this prediction, producing highly value for classified state. combined approach enhances accuracy while reducing computational complexity, making it suitable real-time applications. Empirical results validate effectiveness demonstrating improved reduced mean-squared error (MSE) mean absolute (MAE). highlights benefit combining classification leverage their complementary strengths more granular predictions. proposed method significantly management performance in diverse environments. By merging regression, adds precision subtlety providing notable advancement field. improves both efficiency reliability marking relevant contribution optimization.

Language: Английский

---

An Effective Virtual Machine Allocation in Federated Cloud by PARAMR-DNN Technique

Communications in computer and information science, Journal Year: 2024, Volume and Issue: unknown, P. 257 - 277

Published: Jan. 1, 2024

Language: Английский

---

Adaptive virtual machine placement: a dynamic approach for energy-efficiency, QoS enhancement, and security optimization

Cluster Computing, Journal Year: 2024, Volume and Issue: 28(1)

Published: Oct. 23, 2024

Language: Английский

---