A systematic review on effective energy utilization management strategies in cloud data centers

Journal of Cloud Computing Advances Systems and Applications, Journal Year: 2022, Volume and Issue: 11(1)

Published: Dec. 17, 2022

Abstract Data centers are becoming considerably more significant and energy-intensive due to the exponential growth of cloud computing. Cloud computing allows people access computer resources on demand. It provides amenities pay-as-you-go basis across data center locations spread over world. Consequently, consume a lot electricity leave proportional carbon impact environment. There is need investigate efficient energy-saving approaches reduce massive energy usage in servers. This review paper focuses identifying research done field consumption (EC) using different techniques machine learning, heuristics, metaheuristics, statistical methods. Host CPU utilization prediction, underload/overload detection, virtual selection, migration, placement have been performed manage achieve utilization. In this review, savings achieved by compared. Many researchers tried various methods service level agreement violations (SLAV) centers. By heuristic approach, saved 5.4% 90% with their proposed compared existing Similarly, metaheuristic from 7.68% 97%, learning 1.6% 88.5%, 84% when benchmark for variety settings parameters. So, making use could cut down air pollution, greenhouse gas (GHG) emissions, even amount water needed make power. The overall outcome work understand used save

Language: Английский

---

EVCT: An Efficient VM Deployment Algorithm for a Software-Defined Data Center in a Connected and Autonomous Vehicle Environment

IEEE Transactions on Green Communications and Networking, Journal Year: 2022, Volume and Issue: 6(3), P. 1532 - 1542

Published: March 22, 2022

Software-defined data centers (SDDC) are an emerging softwarized model that can monitor the virtual machines' allocation atop cloud servers. SDDC consists of entities like Virtual Machine (VM) and hardware servers connected switches. SDDCs apply VM deployment algorithms to preserve efficient placement processing traffic generated from Connected Autonomous Vehicles (CAV). To enhance user satisfaction, providers always looking for intellectual large-scale incoming traffics, such as Internet Things (IoT) CAV applications, by optimizing service quality level agreement (SLA). This paper is motivated this, raising energy-efficient cluster algorithm named EVCT handle SLA issues in a environment. EVCT leverages similarity between VMs models problem into weighted directed graph. Based on amount VM, adopts "maximum flow minimum cut theory" graph achieve high VMs. The proposed efficiently reduce energy consumption cost, provide services (QoS) users, have good scalability variable workload. We also carried out series experiments use real-world workload evaluate performance EVCT. results illustrate surpasses state-of-the-art terms cost efficiency.

Language: Английский

---

EvoGWP: Predicting Long-Term Changes in Cloud Workloads Using Deep Graph-Evolution Learning

IEEE Transactions on Parallel and Distributed Systems, Journal Year: 2024, Volume and Issue: 35(3), P. 499 - 516

Published: Jan. 23, 2024

Workload prediction plays a crucial role in resource management of large scale cloud datacenters. Although quite number methods/algorithms have been proposed, long-term changes not explicitly identified and considered. Due to shifty user demands, workload re-locations, or other reasons, the ”resource usage pattern” workload, which is usually stable short-term view, may change dynamically range. Such dynamic cause significant accuracy degradation for algorithms. How handle such an open challenging issue. In this paper, we propose Evolution Graph Prediction (EvoGWP), novel method that can predict using delicately designed graph-based evolution learning algorithm. EvoGWP automatically extracts shapelets identify patterns workloads fine-grained level, predicts by considering factors both temporal spatial dimensions. We design two-level importance based shapelet extraction mechanism mine new pattern dimension, graph model fuse interference among different dimension. By combining from each single workloads, then spatio-temporal GNN-based encoder-decoder workloads. Experiments real trace data Alibaba, Tencent Google show improves up 58.6% over state-of-the-art methods. Moreover, outperform methods terms convergence. To best our knowledge, first work identifies accurately

Language: Английский

---

A multivariate transformer-based monitor-analyze-plan-execute (MAPE) autoscaling framework for dynamic resource allocation in cloud environment

Computing, Journal Year: 2025, Volume and Issue: 107(3)

Published: Feb. 13, 2025

Language: Английский

---

A Multidimensional Virtual Resource Allocation Framework With Energy‐Aware Physical Resource Mapping for Green Cloud Computing

Concurrency and Computation Practice and Experience, Journal Year: 2025, Volume and Issue: 37(4-5)

Published: Feb. 28, 2025

ABSTRACT Cloud computing has seen a surge in demand, driven by its scalability and cost efficiency. However, the growing energy consumption of data centers poses significant environmental challenges. This study introduces multidimensional resource allocation model designed to allocate place virtual resources an energy‐efficient manner using combinatorial auction approach. Unlike current approaches, which rely on predefined resources, this allows users request with specific features capacities tailored their workflows. Furthermore, it incorporates flexible bidding language that supports simultaneous requests for multiple logical AND/OR relations. The accommodates various centers, allowing indicate preferred locations. Through optimization problem, identifies most resource‐efficient allocations placements. provides mathematical definition formulation problem. Given complexity explores several heuristic methods, including ant colony genetic algorithms. A test case generator is developed simulate real‐life scenarios. effectiveness proposed solutions assessed through experiments, demonstrating these methods can achieve near‐optimal within reasonable timeframes.

Language: Английский

---