Why You Should Consider Green Cloud Computing?

By CIOReview | Friday, September 16, 2016

No doubt cloud computing is a cost effective and a scalable infrastructure to run any enterprise application, but the increasing power demands to keep these cloud environments up and running at all times has been a cause of concern. Increased energy consumption leads to faster degenerating of the IT hardware such as servers and switches while giving way for increased carbon emissions.

The U.S. Environmental Protection Agency (EPA) reported that in 2010, data centers consumed the energy of about $11.5Bn and usually energy costs in a data center increase to its double in every five years. According to International Data Corporation (IDC), in 2012, if an estimate of about $1 is spent on hardware then the same amount of money is used for providing power and cooling facilities to data centers. Data centers release a large amount of greenhouse gases that leads to ozone depletion. Gartner reported that in 2008, data centers emitted 116Mn tonnes of CO2, which is more than the overall emission of Nigeria. To curtail these threats, it is incumbent on IT decision makers to use those solutions, processes, and best practices that facilitate minimal environmental footprint. Today, many organizations are finding ways to implement eco-friendly cloud infrastructure a.k.a ‘Green Cloud Computing’ with the aim of reducing the energy consumption of data centers while cutting down the CO2 emissions in these environments.

The concept of green cloud computing is conceptualized to tackle issues related to high-energy consumption by building environmentally sustainable and friendly computing centers and data centers. The eco-friendly power systems will reduce power consumption and harmful emissions of CO2. It will reduce environmental footprints of cloud computing by considering energy consumption as the prime motive along with taking account of types of materials used and economy of other resources like water, electricity, etc. Green cloud computing is also defined as the act of optimizing IT resources to minimize its after-effects and environmental footprints by controlling scarce resources and limiting electronic waste after component manufacturing and recycling. Green cloud computing faces the challenge of regulate the use of resources while continuing to fulfill data center requirements, maintaining its service quality and robustness.

Today, data centers running cloud environments host variety of applications that run for few seconds up to longer time-periods on shared hardware platforms. The requirement to manage multiple applications in a data center has created the challenge of on-demand resource allocation and its provisioning, in response to time-varying workloads. Until now, the only concern of data centers has been to provide high performance without paying much attention towards energy consumption. It is estimated that an average data center consumes energy equal to 25,000 houses. The energy costs are increasing in comparison to availability. Therefore, it is necessary to optimize performance of datacenter resources.

Motivation for Green Cloud Computing

Current IT systems rely upon a complex combination of hardware, networks, and workforce and green cloud computing needs to cover all these areas. It also includes considerable revenue motivations for the companies that will help in controlling of its own power consumption. Hence, the solution needs to address regulatory compliance, end-user satisfaction, return on investment (ROI), and management restructuring. The selection of most powerful management tools with ease of use functionality is required and organizations need to implement various approaches to fulfill these requirements. Implementing simple approaches also put hard impact on finding energy-efficient solutions.

Calculation of total Energy Consumption

The energy consumption in cloud environments is calculated by doing an empirical analysis of cloud model and the analysis is done by considering different runtime tasks. Each task needs to be considered as a unit and the energy produced after each task will be measured on various configurations. During the calculation of the energy consumption, each task needs to be analyzed on various parameters including the size of data to be processed, number of processes, system configuration, and the size of transmitting data. It will help in identification of the relationship between energy consumption and the tasks running in the cloud environments along with the performance and system configuration. The results of analysis correlate energy consumption with system performance, which is important for developing energy efficient mechanisms.

Workload Scheduling

Optimizing the scheduling of workloads across the servers is key to rein in the operating costs. The prime focus of workload scheduling should be to achieve maximum utilization in a cost-effective manner. An approach can be decided by implementing queuing theory principles along with the relationship between service rate, response time, and packet arrival rate. The optimal efficiency can be achieved by maintaining server configurations according to the workload requirements.

Efficient Server Allocation

The efficient management of network resources can be performed by allocating workloads according to the functional capacity of application traffic that will increase the speed of the server. Efficiency will be achieved by minimizing the packet loss and efficiently utilizing the server capacity according to the traffic patterns. The highest optimization level can be achieved by selecting servers, which can process workloads at the matching speed of packet data arrival rates.

Terminal Servers

Terminal servers can also use as an active method for green computing. The Aqua Connect Terminal Server for Mac and Terminal Services for Windows delivers operating systems to the end users. While using the operating systems, the terminal users can connect to terminal servers and the whole of the computing is performed on the server, but the end users will experience the OS on the terminal. The use of terminal servers along with thin clients is increased to create virtual labs. Thin clients will use up to one-eighth of the energy compared to normal workstations. The combination of thin clients and terminal servers provides easy access of operating systems to the end users that reduce energy costs and consumption.

The efficient and effective utilization of computational resources in the cloud will help ineffectively achieving the goals of Green Cloud Computing. Virtualization helps in better utilization of the resources in the cloud environment. The scheduling and migration of virtual machines is important, but monitoring the amount of power consumption and cost included in the migration process is also necessary because it helps in evaluation of system performances.

The important aspect of green cloud computing is to make equipment energy-efficient and eco-friendly. Cloud providers can decrease energy demand by using renewable energy sources. Green cloud computing can be seen as the next generation technology that will help in reducing power usage while adressing the environmental concerns effectively.