A Taxonomy and Future Directions for Sustainable Cloud Computing: 360 Degree View

A Taxonomy and Future Directions for Sustainable Cloud Computing: 360 Degree View
SUKHPAL SINGH GILL and RAJKUMAR BUYYA, The University of Melbourne, Australia
The cloud computing paradigm offers on-demand services over the Internet and supports a wide variety of applications. With the recent growth of Internet of Things (IoT) based applications the usage of cloud services is increasing exponentially. The next generation of cloud computing must be energy-efficient and sustainable to fulfil the end-user requirements which are changing dynamically. Presently, cloud providers are facing challenges to ensure the energy efficiency and sustainability of their services. The usage of large number of cloud datacenters increases cost as well as carbon footprints, which further effects the sustainability of cloud services. In this paper, we propose a comprehensive taxonomy of sustainable cloud computing. The taxonomy is used to investigate the existing techniques for sustainability that need careful attention and investigation as proposed by several academic and industry groups. Further, the current research on sustainable cloud computing is organized into several categories: application design, sustainability metrics, capacity planning, energy management, virtualization, thermal-aware scheduling, cooling management, renewable energy and waste heat utilization. The existing techniques have been compared and categorized based on the common characteristics and properties. A conceptual model for sustainable cloud computing has been proposed along with discussion on future research directions. CCS Concepts: A.1 [General Literature]: Introductory and Survey; C.0 [General]: Systems Architectures; C.2.4 [Computer- Communication Networks]: Distributed Systems; D.4.1 [Process Management]: Scheduling; H.3.4 [Systems and Software]: Distributed Systems; J.7 [Distributed Parallel and Cluster Computing]; K.6.2 [Management of Computing and Information Systems]: Installation Management General Terms: Documentation, Cloud Computing, Methodical Analysis, Conceptual Model, Focus of Study, Research Challenges, Theory, Reference Architecture, Trade-off, Future Directions, Management, Systematic Review, Survey
Additional Key Words and Phrases: Energy-efficiency, Sustainability, Cloud Datacenters, Quality of Service, Green Computing, Holistic Management, Sustainable Cloud Computing, Application Design, Energy Management, Renewable Energy, Thermal-Aware Scheduling, Virtualization, Sustainable Cloud Datacenters, Capacity Planning, Sustainable Metrics, Cooling Management and Waste Heat Utilization

1. INTRODUCTION
   Cloud computing offers a flexible and powerful computing environment to provide on-demand, subscription-based online services over the Internet to host applications on a pay-as-you-go basis. The various cloud providers such as Microsoft, Google and Amazon make extensive use of Cloud Data Centers (CDCs) to fulfill the requirements (memory, data, compute or network) of the digital world. To reduce the service delay and maintain the Service Level Agreement (SLA), fault tolerance should be provided through replicating the compute abilities redundantly [1]. To ensure the availability and reliability of services, the components of CDCs such as network devices, storage devices and servers should be run 24/7 [2]. Large amounts of data are created by digital activities such as data streaming, file sharing, searching and social networking websites, e-commerce, sensor networks and that data can be stored as well as processed efficiently using CDCs [3] [4]. The energy cost is added by creating, processing and storing each bit of data, which increases carbon footprints that further impacts on the sustainability of cloud services. Due to the large consumption of electricity by CDCs, the research community is addressing challenge of designing sustainable CDCs [5].
   With the continuous growth of Internet of Things (IoT) based applications, the usage of cloud services is increasing exponentially, which further increases the electricity consumptions of CDCs by 20-25% every year [6]. Existing studies claimed that 78.7 million metric tons of CO2 are produced by datacenters, which is equal to the two percent of global emissions [7]. CDCs in USA consumed 100 billion kilowatt hours (kWh) in 2015, which is sufficient for Washington City [11]. The consumption of electricity will reach 150 billion kWh by 2022 i.e. increase by 50% [12]. Energy consumption in CDCs can be increased to 8000 terawatt hours (TWh) in 2030 if controlled mechanisms are not identified [122]. Due to underloading and overloading of resources in infrastructure (cooling, computing, storage, networking etc.), the energy consumption in cloud datacenters is not efficient and mostly the energy is consumed while some of the resources are in idle state, which increases the cost of cloud services [11]. Figure 1

…(Full text truncated)…