Predictive VM placement algorithm for resource optimization: leveraging deep learning forecasting and resource relationship modeling

International Journal of Computers and Applications, Год журнала: 2025, Номер unknown, С. 1 - 14

Опубликована: Фев. 28, 2025

The growing demand for cloud computing has made it imperative to optimize the utilization of resources. Resource optimization can be improved through Virtual Machine (VM) Placement. In order effectively placement VMs, becomes necessary anticipate future resource demand. However, accurate forecasting is a major challenge due dynamic nature applications. Furthermore, if VMs are placed on same server, lead contention, especially when they compete This contention adversely affect performance and potentially increase cost users, as well energy consumption by infrastructure. work proposes model named Predictive Disparity-based Placement (PDVMP) which aims enhance VM decision. integrates techniques grounded in Deep Learning estimating needs VMs. estimation incorporated decision ensure long-term sustainability destination server. Moreover, used current research balances execution packing multiple that exhibit complementary physical PDVMP tested against benchmark policies using real workload traces bitBrains datacenter. results show proposed approach improves while reducing both bottlenecks consumption. experimentation shows an improvement Energy Performance Metric ranging from 49.3% 62.97%.

Язык: Английский

Optimizing Energy Efficiency in Cloud Data Centers: A Reinforcement Learning-Based Virtual Machine Placement Strategy

Network, Год журнала: 2025, Номер 5(2), С. 17 - 17

Опубликована: Май 27, 2025

Cloud computing faces growing challenges in energy consumption due to the increasing demand for services and resource usage data centers. To address this issue, we propose a novel energy-efficient virtual machine (VM) placement strategy that integrates reinforcement learning (Q-learning), Firefly optimization algorithm, VM sensitivity classification model based on random forest self-organizing map. The proposed method, RLVMP, classifies VMs as sensitive or insensitive dynamically allocates resources minimize while ensuring compliance with service level agreements (SLAs). Experimental results using CloudSim simulator, adapted from Microsoft Azure, show our significantly reduces consumption. Specifically, under lr_1.2_mmt strategy, achieves 5.4% reduction compared PABFD, 12.8% PSO, 12% genetic algorithms. Under iqr_1.5_mc reductions are even more significant: 12.11% 15.6% 18.67% Furthermore, number of live migrations, which helps SLA violations. Overall, combination Q-learning algorithm enables adaptive, SLA-compliant improved efficiency.

Язык: Английский

RAP-Optimizer: Resource-Aware Predictive Model for Cost Optimization of Cloud AIaaS Applications

Electronics, Год журнала: 2024, Номер 13(22), С. 4462 - 4462

Опубликована: Ноя. 14, 2024

Artificial Intelligence (AI) applications are rapidly growing, and more joining the market competition. As a result, AI-as-a-service (AIaaS) model is experiencing rapid growth. Many of these AIaaS-based not properly optimized initially. Once they start large volume traffic, different challenges revealing themselves. One maintaining profit margin for sustainability AIaaS application-based business model, which depends on proper utilization computing resources. This paper introduces resource award predictive (RAP) cost optimization called RAP-Optimizer. It developed by combining deep neural network (DNN) with simulated annealing algorithm. designed to reduce underutilization minimize number active hosts in cloud environments. dynamically allocates resources handles API requests efficiently. The RAP-Optimizer reduces physical an average 5 per day, leading 45% decrease server costs. impact was observed over 12-month period. observational data show significant improvement utilization. effectively operational costs from USD 2600 1250 month. Furthermore, increases 179%, 600 1675 inclusion dynamic dropout control (DDC) algorithm DNN training process mitigates overfitting, achieving 97.48% validation accuracy loss 2.82%. These results indicate that enhances management cost-efficiency applications, making it valuable solution modern

Язык: Английский