Energy-driven cloud simulation: existing surveys, simulation supports, impacts and challenges
- PDF / 453,377 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 99 Downloads / 197 Views
(0123456789().,-volV)(0123456789(). ,- volV)
Energy-driven cloud simulation: existing surveys, simulation supports, impacts and challenges Azlan Ismail1 Received: 4 August 2019 / Revised: 17 December 2019 / Accepted: 1 February 2020 Ó Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract A large scale cloud data center is needed to provision various applications in different domains. As a result, power consumption is expected to increase due to huge operations and expansion of cloud data centers. Furthermore, it also intensifies environment concern. Various approaches and solutions for energy-driven cloud data center have been proposed to overcome this challenge. Testing and evaluating these solutions in large scale is costly and time consuming. Hence, simulation techniques become the preferred approach to tackle this concern. There are a few cloud simulators have been developed with different features and capabilities which can be chosen for this reason. A survey work can serve as a guideline. A few cloud simulation surveys have been done but limited survey is found for energy-driven cloud simulation. This review complements the existing surveys by considering different aspects of energy-driven cloud simulators. Therefore, this paper presents a review of existing cloud simulation surveys with several classifications. Furthermore, it provides some insights of the selected cloud simulators by emphasizing on the energy-driven simulation supports and the impact of the cloud simulators in succeeding works. This paper also highlights open and future challenges. Keywords Cloud simulator Energy efficiency Energy metrics Power management
1 Introduction Cloud data centers are growing exponentially in the number of hosted servers to connect hundreds of thousands of servers [6]. The growth is significant to run a wide range of applications in various domains, such as scientific applications, healthcare, e-commerce, smart grids and nuclear science [7]. Data centers being large scale computing infrastructures require huge energy budgets to operate properly and to cool down the heat generated, which results in various energy efficiency issues [16, 43, 73]. Furthermore, the enormous green house gases (GHG) emissions produced by a large scale data center have intensified environment concerns [7]. Hence, various effort and solutions have been proposed
& Azlan Ismail [email protected] 1
Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA (UiTM), 40450 Shah Alam, Selangor, Malaysia
and implemented to address these issues. In particular, energy efficiency within data centers can be achieved by exploiting workload consolidation, energy-aware workload placement, and proportional computing [7]. Providing and deploying energy efficient related solutions is a non-trivial task for the cloud providers. They need efficient tools to support solution design decisions related to deployment models, resource allocation, scheduling, and performance adjustments. Evaluating solutions direct
Data Loading...
