Article
Computer Science, Information Systems
Avinab Marahatta, Sandeep Pirbhulal, Fa Zhang, Reza M. Parizi, Kim-Kwang Raymond Choo, Zhiyong Liu
Summary: With the rapid growth of cloud data centers (CDCs) due to the popularity of cloud computing and high-performance computing, the need to maximize resource utilization and ensure energy efficiency has become critical. An energy-efficient dynamic scheduling scheme (EDS) for real-time tasks in virtualized CDCs has been proposed in this paper to address the challenges of inefficient resource utilization and high energy consumption. By classifying and merging tasks based on historical scheduling records, the EDS significantly improves overall scheduling performance, increases CDC resource utilization, and reduces energy consumption by utilizing the energy efficiencies and optimal operating frequencies of heterogeneous physical hosts.
IEEE TRANSACTIONS ON CLOUD COMPUTING
(2021)
Article
Computer Science, Information Systems
Kalka Dubey, S. C. Sharma
Summary: Cloud computing offers useful services but also brings security risks to user information privacy. This study proposes an extended intelligent water drop algorithm and a VM allocation algorithm to optimize task execution in a secure cloud environment. Experimental results show the effectiveness of the proposed algorithm compared to existing approaches.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Anurina Tarafdar, Soumi Sarkar, Rajib K. Das, Sunirmal Khatua
Summary: An accurate host power model is crucial for effective power management in data centers, which helps reduce energy consumption and cost. Researchers have analyzed existing power models and proposed three power models based on multi-variable Linear Regression, Support Vector Regression (SVR), and Artificial Neural Network (ANN). Experimental results show that the proposed power models, especially those based on SVR and ANN, accurately predict host power consumption.
JOURNAL OF GRID COMPUTING
(2023)
Article
Computer Science, Information Systems
Neeraj Kumar Sharma, Sriramulu Bojjagani, Y. C. A. Padmanabha Reddy, Manojkumar Vivekanandan, Jagadeesan Srinivasan, Anup Kumar Maurya
Summary: Due to the rapid utilization of cloud services, the energy consumption of cloud data centres is increasing dramatically. This paper proposes a Branch-and-Price based energy-efficient VMs allocation algorithm and a Multi-Dimensional Virtual Machine Migration (MDVMM) algorithm at the cloud data center. The experimental results demonstrate that these algorithms significantly reduce energy consumption and improve resource utilization.
Article
Computer Science, Artificial Intelligence
Ammar Al-Moalmi, Juan Luo, Ahmad Salah, Kenli Li, Luxiu Yin
Summary: This paper studies the container and VM placement problem in CaaS environments and proposes an algorithm based on the Whale Optimization Algorithm to optimize power consumption and resource utilization. The proposed method outperforms existing methods in experimental evaluations.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Thermodynamics
Elham Hormozi, Shuwen Hu, Zhe Ding, Yu-Chu Tian, You-Gan Wang, Zu-Guo Yu, Weizhe Zhang
Summary: Energy efficiency is a critical issue in data centre management. This study focuses on using an accelerated Genetic Algorithm (GA) to optimize the placement of virtual machines (VMs) in data centres. The simulation results show that the accelerated GA is faster and provides better solutions compared to the standard GA.
Article
Computer Science, Information Systems
Hong-Yen Lo, Wanjiun Liao
Summary: The study focuses on survivability of virtual data centers and proposes the CALM algorithm to minimize network resource usage and ensure survivability after hardware failures.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2021)
Article
Computer Science, Hardware & Architecture
Ali Aghasi, Kamal Jamshidi, Ali Bohlooli, Bahman Javadi
Summary: The traditional method of saving energy in Virtual Machine Placement (VMP) by consolidating more VMs in fewer servers can lead to server overheating and performance degradation. The lack of an accurate and efficient model for the data center environment makes it challenging to develop an effective VMP mechanism. Data-driven approaches have limitations due to insufficient data and system changes. Researchers have turned to model-free paradigms like reinforcement learning, but scalability and exploration costs are major challenges. This paper presents a decentralized implementation of reinforcement learning for VMP in data centers, achieving energy consumption reduction and temperature control while meeting SLAs. Experimental results show significant improvements compared to baseline algorithms, including accelerated policy convergence after configuration changes.
Article
Computer Science, Information Systems
Shikha Mehta, Parmeet Kaur, Parul Agarwal
Summary: The Infrastructure as a Service (IaaS) model of cloud computing offers cost-efficient resources in the form of virtual machines (VM) by utilizing cloud-based physical machines. The paper focuses on the VM placement problem, aiming to optimize the mapping of VMs to physical machines to reduce costs and energy consumption. Two variants of the whale optimization algorithm (WOA) are proposed, with improved exploitation and exploration phases. Experimental results show that these variants provide efficient solutions for different workloads while reducing computational complexity.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Telecommunications
Zhou Zhou, Mohammad Shojafar, Mamoun Alazab, Jemal Abawajy, Fangmin Li
Summary: Cloud Data Centers (CDCs) are significant for enterprises, but the increased demand for computing power, especially for IoT applications, leads to substantial energy consumption. Existing green resource allocation algorithms focus on minimizing the number of active Physical Machines (PMs) and lack a comprehensive consideration of the energy efficiency of Virtual Machines (VMs) and load fluctuations. Our proposed adaptive energy-aware VM allocation and deployment mechanism efficiently handles load fluctuations and demonstrates superior performance in energy consumption and Service Level Agreements (SLA) violation.
IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING
(2021)
Article
Computer Science, Hardware & Architecture
Hao Feng, Yuhui Deng, Jie Li
Summary: This study proposes a global-energy-aware virtual machine placement strategy to reduce the total energy consumption of data centers, and designs a two-step SAG algorithm to lower the energy consumption of cloud data centers with multiple deployed VMs. Experimental results show that compared with other algorithms, this strategy can reduce the total energy consumption of cloud data centers by 8%-24.9%.
JOURNAL OF SYSTEMS ARCHITECTURE
(2021)
Article
Computer Science, Information Systems
Suruchi Talwani, Jimmy Singla, Gauri Mathur, Navneet Malik, N. Z. Jhanjhi, Mehedi Masud, Sultan Aljahdali
Summary: This paper introduces a machine-learning-based approach for dynamically integrating virtual machines (VMs) to meet the standards of service level agreements (SLAs). By predicting usage thresholds adaptively, it improves resource usage and enhances performance.
Article
Computer Science, Hardware & Architecture
Xingxing Li, Weidong Li, Xuejie Zhang
Summary: We propose a new mechanism called dominant resource fairness with time window constraints (DRFTW) to solve the problem of multiresource fair allocation with time window constraints for user tasks in cloud computing systems. We design an algorithm for DRFTW to achieve the optimal allocations. DRFTW has desirable properties such as no user can improve its performance without making others worse off, no user prefers equal allocations, no user envies others' allocations, and no user can benefit from providing false information. Simulations based on Alibaba Cluster Trace show that DRFTW significantly increases the minimum dominant share and improves fairness compared to conventional fair allocation methods.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Information Systems
Xiaoqi Zhang, Hongju Cheng, Zhiyong Yu, Neal N. Xiong
Summary: In this article, a multiresource allocation system for cooperative computing in the Internet of Things based on deep reinforcement learning is proposed. By redefining calculation models and considering practical interference factors, the system efficiently supports complex applications. Experiments have shown that this system has low service latency under resource-constrained conditions, and the improvement is more significant with the increase of network size.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Chemistry, Multidisciplinary
Nimra Malik, Muhammad Sardaraz, Muhammad Tahir, Babar Shah, Gohar Ali, Fernando Moreira
Summary: This article addresses the issue of energy consumption and efficient resource utilization in virtualized cloud data centers, proposing an algorithm based on task classification and thresholds for efficient scheduling. Experiments validate the effectiveness of the proposed technique over other algorithms in terms of energy consumption, makespan, and load balancing.
APPLIED SCIENCES-BASEL
(2021)