IUP Publications Online
Home About IUP Magazines Journals Books Archives
     
Recommend    |    Subscriber Services    |    Feedback    |     Subscribe Online
 
The IUP Journal of Information Technology
Energy-Aware Simulators for Efficient Data Center Design
:
:
:
:
:
:
:
:
:
 
 
 
 
 
 

With the growing popularity of cloud computing, data centers are becoming very much important for the provisioning of computing resources. Data centersí cost and operating expenses have increased with the increase in computing capacity. Research has identified that the energy utilized by computing and communication units of data center contributes a considerable portion of the data center operational costs. Setting up and running these experiments in real cloud environments are very much costly to optimize the designs. However, modeling and simulation tools are suitable solutions that often provide good alternatives for emulating cloud computing environments. Several simulation tools have been developed to provide a simulation environment for energy-efficient cloud computing data centers. A comprehensive review has been presented for the data center simulators and a comparative analysis has been carried out to outline the strengths. The work is very much helpful in selecting the simulators for efficient data center design.

 
 

Smart computing devices have become most popular in the recent few years. These devices have proved to be useful, providing tools that are ready anywhere and anytime, to serve a userís needs. Many of these services are delivered using a cloud computing environment, in which the resources are retrieved from internet over the web-based applications and tools, in place of a direct connection to the server. Hence, data is stored in servers (Bugiel et al., 2011). Cloud computing model enables access to information, provided an electronic device has access to the web. Such type of system enables users to work remotely. It allows hosting of applications from scientific, customer and business domains (Buyya et al., 2010; and Singh and Kumar, 2014).

Recent data centers functioning under the cloud computing structure are hosting various applications ranging from those that run for a few seconds to those that run for longer periods of time on communal hardware platforms. Requirement to manage multiple applications in a data center creates the challenge of on-demand resource provisioning and allocation in response to time-varying workloads. Generally, data center resources are allocated to the applications depending upon peak load characteristics, maintaining isolation and providing a guaranteed performance. Therefore, data centers are not only exclusive to maintain but also unfriendly to the environment (Beloglazov et al., 2011). They drive more in carbon emissions. Huge carbon footprints and high energy costs are incurred due to massive amounts of electrical energy required to power and cool several servers hosted in these data centers. Cloud service providers need to adopt measures to make sure that the profit margin is not dramatically reduced because of high energy costs.

Recently, high performance has been the sole concern in data center deployments and this request has been fulfilled without paying much attention to energy consumption. Consistent with world energy outlook, the reported electrical energy consumption is set to rise around 76% for the next 15-16 years with data centers (Rao and Premchand, 2016). As energy costs are increasing while availability dwindles, there is a need to shift the focus from optimizing data center resource management for pure performance to optimizing for energy efficiency while maintaining high service level performance. Therefore, experts are still challenged with how to effectively and efficiently make the data center. Data centers have been estimated to contribute 30-40% of organizationsí energy consumption and greenhouse gas emissions. It is predictable that with the explosion of cloud computing, applications, Internet and networking, data centers could contribute 9% to technology Green House Gas (GHG) emissions and 4% of this could be from cooling by 2020 (https://www.sustainablebusinesstoolkit.com/data-center-efficiency/). The cooling process accounts for 40% of all power consumed by data centers, so this issue is top of mind for operators (Lata and Kumar, 2016)). Ensuring optimum energy-efficient data center not only lowers operational expenditure but reduces the strain on equipment energy efficiency mechanisms, extending the lifespan of the hardware, freeing up power for IT equipment and increasing equipment uptime. The decision to invest in most efficient infrastructure is easy, however, choosing the method with which they regulate temperature within the data center can be more challenging.

Energy consumption has become an important problem for large-scale data centers, because of high and constant power demand. This necessitates research for dynamically reducing the amount of energy used for cooling, computing and maintaining a data center. Primary motivations have always been to develop data center power management tools that deliver energy efficiency, while minimizing the impact on performance. Consequently, energy-efficient data center design and management has been a challenge of increasing importance. Energy consumption in data centers can be reduced by efficient data centers design, efficient management of computing resources and cooling units (Pedram, 2012). Foremost difficulty in the data center design is the lack of a holistic simulator, where the impact of innovative computing resource management techniques can be tested with different designs of data centers. This in turn requires the designers to be proficiently aware of the domain-specific design tools that require user intervention design process in each step. The data center energy-efficient simulators can be very much effective for the design and enhancement of energy-efficient data centers. Data center simulator can be used for studying the energy efficiency of data centers under numerous data center geometries, platform power management schemes, scheduling algorithms and workload characteristics. Therefore, a suitable alternative is the use of models and simulation tools to emulate cloud environments and conduct the necessary experimentation. CloudSim, ECOFEN, GDCSim, MDCSim, DCSim, GreenCloud and iCanCloud are some of the most powerful simulation tools for cloud computing environments which support researchers to model multiple complex scenarios (Buyya et al., 2009; Gupta

et al., 2011; Sinha and Shekhar, 2015; and Son et al., 2015). These tools are used to iteratively design the energy-efficient data centers

 
 

Information Technology Journal,Cloud computing, Data center design, Energy aware simulator.