Cloud technology is often called ‘green’ because it uses computing infrastructure more efficiently. But green doesn’t refer to process optimization, rather it refers to the environmental impact of any system. Businesses outsourcing operations to the cloud may not be reducing their carbon footprint, they are just moving it elsewhere. Calculating the difference between energy consumption of cloud data centers and on-site resources is also not easy.
Most data centers are large facilities and consume a lot of power. Depending on the source of electricity, they might be creating vast amounts of pollution. However, as more of them are being built, companies are also realizing the need to reduce energy consumption. Many techniques have been developed and utilized to deliver improved resource efficiency as well as energy efficiency.
Cloud infrastructure can be greener than multiple, small, corporate IT environments for the following reasons:
1. Server Virtualization
Virtualization is a technology that allows a single physical server to run multiple operating system images at the same time. They exist independently of each other and in most cases are not even aware of the presence of other OSes. The advantages of this technique are twofold. The first is that the resources of one server are used to its maximum potential, since running only one application or OS often leaves the processor idle for long periods of time. The other is that less physical equipment is needed to the run same workload. This in turn leads to lower power consumption and less electronic waste at the end of the servers’ lifecycle.
2. Multitenancy
Multitenancy allows different organizations or various departments within the same organization to use common cloud based infrastructure. Small companies, which can have very inefficient IT resources, are able to benefit from the efficiencies of scale offered by large data centers. By combining demand patterns of different companies, cloud providers are able to even out the fluctuations between peak and average computing usage. Again, this eliminates the need for multiple companies to have equipment to deal with individual peak demand.
3. Pay-per-use model
Cloud computing runs on the pay per use or subscription business model. Each business pays for the resources it uses, encouraging them to better manage their utilization. Users can also access extra computing power only when they need it, instead of buying hardware which remains idle for the majority of time.
4. Green power sources
Concentrating computing resources in one data center also offers a hidden advantage. Persuading a large number of companies with their separate servers to switch from ‘dirty’ power sources to renewable energy is not easy, since the individual costs may be too high, especially for small and medium businesses. It is easier for one data center to change and use clean energy such as solar arrays or wind energy. This means a large portion of servers can reduce their power consumption at one stroke, instead of waiting for individual companies to catch up.
Many large companies like Google, Microsoft and Amazon are also making continuous improvements to their data centers, which can be hard for smaller organizations to match. All of these optimizations result in a shared data center saving energy compared to a corporate IT department.
About the Author: Chris Smith is a leading cloud technologist and creator of Best Cloud Hosting comparison website.
License: Image author owned
