Elastic Load Balancing for Dynamic Virtual Machine Reconfiguration Based on Vertical and Horizontal Scaling

Drastic variations in high-performance computing workloads lead to the commencement of large number of datacenters. To revolutionize themselves as green datacenters, these data centers are assured to reduce their energy consumption without compromising the performance. The energy consumption of the processor is considered as an important metric for power reduction in servers as it accounts to 60% of the total power consumption. In this research work, a power-aware algorithm (PA) and an adaptive harmony search algorithm (AHSA) are proposed for the placement of reserved virtual machines in the datacenters to reduce the power consumption of servers. Modification of the standard harmony search algorithm is inevitable to suit this specific problem with varying global search space in each allocation interval. A task distribution algorithm is also proposed to distribute and balance the workload among the servers to evade over-utilization of servers which is unique of its kind against traditional virtual machine consolidation approaches that intend to restrain the number of powered on servers to the minimum as possible. Different policies for overload host selection and virtual machine selection are discussed for load balancing. The observations endorse that the AHSA outperforms, and yields better results towards the objective than, the PA algorithm and the existing counterparts.

Download Full-text

Virtual machine scheduling strategy based on machine learning algorithms for load balancing

EURASIP Journal on Wireless Communications and Networking ◽

10.1186/s13638-019-1454-9 ◽

2019 ◽

Vol 2019 (1) ◽

Cited By ~ 2

Author(s):

Xin Sui ◽

Dan Liu ◽

Li Li ◽

Huan Wang ◽

Hongwei Yang

Keyword(s):

Machine Learning ◽

Load Balancing ◽

Virtual Machine ◽

Learning Algorithms ◽

Machine Scheduling ◽

Machine Learning Algorithms ◽

Scheduling Strategy ◽

Virtual Machine Scheduling

Download Full-text

Efficient task scheduling on virtual machine in cloud computing environment

International Journal of Pervasive Computing and Communications ◽

10.1108/ijpcc-04-2020-0029 ◽

2020 ◽

Vol ahead-of-print (ahead-of-print) ◽

Author(s):

Mahfooz Alam ◽

Mahak ◽

Raza Abbas Haidri ◽

Dileep Kumar Yadav

Keyword(s):

Cloud Computing ◽

Load Balancing ◽

Response Time ◽

Virtual Machine ◽

Task Allocation ◽

Completion Time ◽

Virtual Machines ◽

Scheduling Algorithm ◽

Content Type ◽

Minimum Completion Time

Purpose Cloud users can access services at anytime from anywhere in the world. On average, Google now processes more than 40,000 searches every second, which is approximately 3.5 billion searches per day. The diverse and vast amounts of data are generated with the development of next-generation information technologies such as cryptocurrency, internet of things and big data. To execute such applications, it is needed to design an efficient scheduling algorithm that considers the quality of service parameters like utilization, makespan and response time. Therefore, this paper aims to propose a novel Efficient Static Task Allocation (ESTA) algorithm, which optimizes average utilization. Design/methodology/approach Cloud computing provides resources such as virtual machine, network, storage, etc. over the internet. Cloud computing follows the pay-per-use billing model. To achieve efficient task allocation, scheduling algorithm problems should be interacted and tackled through efficient task distribution on the resources. The methodology of ESTA algorithm is based on minimum completion time approach. ESTA intelligently maps the batch of independent tasks (cloudlets) on heterogeneous virtual machines and optimizes their utilization in infrastructure as a service cloud computing. Findings To evaluate the performance of ESTA, the simulation study is compared with Min-Min, load balancing strategy with migration cost, Longest job in the fastest resource-shortest job in the fastest resource, sufferage, minimum completion time (MCT), minimum execution time and opportunistic load balancing on account of makespan, utilization and response time. Originality/value The simulation result reveals that the ESTA algorithm consistently superior performs under varying of batch independent of cloudlets and the number of virtual machines’ test conditions.

Download Full-text