Elsevier

Information Sciences

Volume 258, 10 February 2014, Pages 452-462
Information Sciences

Optimizing Energy Consumption with Task Consolidation in Clouds

https://doi.org/10.1016/j.ins.2012.10.041Get rights and content

Abstract

Task consolidation is a way to maximize utilization of cloud computing resources. Maximizing resource utilization provides various benefits such as the rationalization of maintenance, IT service customization, QoS and reliable services, etc. However, maximizing resource utilization does not mean efficient energy use. Much of the literature shows that energy consumption and resource utilization in clouds are highly coupled. Consequently, some of the literature aims to decrease resource utilization in order to save energy, while others try to reach a balance between resource utilization and energy consumption. In this paper, we present an energy-aware task consolidation (ETC) technique that minimizes energy consumption. ETC achieves this by restricting CPU use below a specified peak threshold. ETC does this by consolidating tasks amongst virtual clusters. In addition, the energy cost model considers network latency when a task migrates to another virtual cluster. To evaluate the performance of ETC we compare it against MaxUtil. MaxUtil is a recently developed greedy algorithm that aims to maximize cloud computing resources. The simulation results show that ETC can significantly reduce power consumption in a cloud system, with 17% improvement over MaxUtil.

Introduction

Cloud computing has recently become popular due to the maturity of related technologies such as network devices, software applications and hardware capacities. Resources in these systems can be widely distributed and the scale of resources involved can range from several servers to an entire data center. To integrate and make good use of resources at various scales, cloud computing needs efficient methods to manage them [4]. Consequently, the focus of much research in recent years has been on how to utilize resources and how to reduce power consumption.

One of the key technologies in cloud computing is virtualization. The ability to create virtual machines (VMs) [14] dynamically on demand is a popular solution for managing resources on physical machines. Therefore, many methods [17], [18] have been developed that enhance resource utilization such as memory compression, request discrimination, defining threshold for resource usage and task allocation among VMs. Improvements in power consumption, and the relationship between resource usage and energy consumption has also been widely studied [6], [10], [11], [12], [14], [15], [16], [17], [18]. Some research aims to improve resource utilization while others aim to reduce energy consumption. The goals of both are to reduce costs for data centers. Due to the large size of many data centers, the financial savings are substantial.

Energy consumption varies according to CPU utilization [11]. Higher CPU utilization usually implies greater energy consumption. However, higher CPU utilization does not equate to energy efficiency. This phenomenon motivates the idea of not exhausting CPUs with high levels of utilization (for example, 80%∼100%) in order to save energy. To this end, we propose an energy-aware task consolidation (ETC) method that minimizes energy consumption.

The main idea of ETC is to ration CPU utilization and manage task consolidation amongst virtual clusters. In addition, our energy cost model considers network latency when a task migrates to another virtual cluster. The main contributions of our work are as follows. First, we present a method to ration CPU utilization and manage task consolidation amongst virtual clusters. Secondly, we show how ETC can reduce power consumption significantly by managing task consolidation in a cloud system. Third we compare our results to a recent greedy method called MaxUtil[10] that attempts to reduce energy consumption by assigning as many tasks as it can to a VM.

The rest of this paper is organized as follows. Section 2 explains our research model. Section 3 presents the proposed techniques on task consolidation and energy saving. In Section 4, the simulation results and performance analysis are given to weigh the pros and cons of the proposed method. In Section 5, we discuss related work. Finally, the conclusion and future work are presented in Section 6.

Section snippets

Research Model

The research model for this study is presented in Fig. 1, which shows a cloud system that consists of several virtual clusters (VC). Each virtual cluster provides a limited number of VMs. Without losing generality, VMs are used as a basic unit to execute a task. The percentage of CPU utilization is used to judge whether a VM has enough resources available for a service. Fig. 1(b) gives an example of network bandwidth between virtual clusters, which are geographically distributed. This shows

Energy Efficient Task Consolidation

In this section, we present an energy-aware task consolidation (ETC) method to optimize energy usage in cloud systems. In the energy model presented in Fig. 2(b), a VM is assumed to consume α watt/s in its idle state. An additional β watt/s is required for executing tasks when CPU utilization is between 0% and 20%. If CPU utilization is between 20% and 50%, the additional energy consumed increases to 3β watt/s. Energy is consumed at a greater rate as CPU utilization increases. For instance,

Experiment Configuration

To evaluate the performance of the proposed technique, we implemented the ETC method and the MaxUtil [10] method. The MaxUtil method consolidates tasks and assigns as many tasks as it can to a VM. Overall, MaxUtil is the same as ETC except it has a 100% CPU utilization threshold. The cloud computing model in this section is the same as we described in Fig. 1 but for the purpose of analysis, the numbers of nodes in the virtual clusters are different. In our test, we change the number of nodes in

Related Work

Energy consumption is an important issue in many fields of research. Both consumers and industries want their products to use less power in order to reduce energy costs. As systems get larger and more complex they typically consume more energy. This problem extends to networks and consequently extends to cloud computing. Since data centers contain large clusters of computers, any reduction in energy expenditure can result in large economical savings. For this reason and others, there has been

Conclusion and Future Work

Clouds typically consist of multiple resources. These resources can be distributed, heterogeneous and virtualized. A high priority for a cloud computing system is the maximization of profits. The amount of energy consumed by these systems has a big influence on how profitable they are. Much of the literature shows energy consumption and resource utilization in clouds are highly coupled. They show that task consolidation is an effective technique to increase resource utilization and that

References (20)

  • Muhammad H Alizai, Georg Kunz, Olaf Landsiedel and Klaus Wehrle, “Promoting Power to a First Class Metric in Network...
  • Jayant Baliga, Robert Ayre, Wayne V. Sorin, Kerry Hinton, Rodney S. Tucker, “Energy Consumption in Access Networks”,...
  • Jayant Baliga, Kerry Hinton, Rodney S. Tucker, “Energy Consumption of the Internet,” Proceedings of Conference on...
  • Andreas. Beri et al.

    Minh Quan Dang and Kostas Pentikousis, “Energy-Efficient Cloud Computing”

    The Computer Journal

    (2010)
  • Philippe Chanclou, Stéphane Gosselin, Juan Fernández Palacios, Victor López Alvarez and Evi Zouganeli, “Overview of the...
  • Xiaobo Fan, Wolf-Dietrich Weber and Luiz Andre Barroso, “Power Provisioning for a Warehouse-sized Computer”,...
  • Chamara Gunaratne, Ken Christensen and Bruce Nordman, “Managing energy consumption costs in desktop pcs and lan...
  • Aman Kansal, Feng Zhao, Jie Liu, Nupur Kothari and Arka A. Bhattacharya, “Virtual Machine Power Metering and...
  • Christoph Lange and Andreas Gladisch, “On the energy consumption of FTTH access networks,” Proceedings of Optical Fiber...
  • Young Choon Lee and Albert Y. Zomaya, “Energy efficient utilization of resources in cloud computing systems”, The...
There are more references available in the full text version of this article.

Cited by (136)

View all citing articles on Scopus
View full text