Optimization of Energy in Data Center
- July 21, 2014
Opt for virtualization is just one of the actions that can be taken to ensure the sustainability of the environment.
A recent study at Stanford University points out that from 2005 to 2013 there was a 48% increase in energy consumption in data centers in the United States. Fewer than many analysts predicted, but still significant. In an economy in constant instability, it is worth taking a look at how to increase energy efficiency in data centers, especially because in many cases the energy efficiency can be obtained without major repairs of IT equipment or facilities. Following up on the drawbacks, necessary changes can be made. Depending on the size of the center, the budget of the facility, and similar other factors, alternative energy options can be adopted like geothermal, hydroelectric, or even uranium energy (generated through Uranium Mining) or a similar option.
That being the case, let’s look at the responsible factors:
1. Determine use
The first step in making the data center more energy efficient is to find out exactly how much electricity the environment uses, and where, specifically, electricity is consumed. My first recommendation is to perform an assessment of the data center to quantify the energy use means for the company. The Green Grid has a metric to use energy called PUE that involves dividing the total energy used in an installation by the IT equipment. You do not need to be sophisticated to quantify the consumption of primary energy. Understanding the PUE can provide a perspective on how you can be more efficient.
2 Check the account
Another very simple way to record improved Energy efficiency is always evaluate carefully the account of the data center, a practice not much performed by IT departments today. It is recommended to examine utility bills in an effort to find areas that could be more energy efficient in the question. While monthly costs do not always correspond to inefficiencies, they can at least highlight where opportunities for improvement are.
3 Prepare for virtualization
Once initial assessments are carried out, examining where to cut electricity usage. Specific strategies will vary from data center to data center. However, server virtualization is a common strategy that can reduce energy consumption and costs significantly by replacing many physical servers. To extract maximum virtualization, or any activity that aims at efficiency, the company has to consider the ripple effect on data center operations. If your server storage is about 80% usage, you must ensure it is monitoring the power requirements to make proper use. By implementing virtualization, the density of workload increases on certain servers, with this, power and cooling should be adjusted according to the racks of servers. A review of the cooling architecture can lead to significant savings with minimal effort. With this action, you can register for energy efficiency around 20% to 40% savings. This action can be as simple as move the air-conditioning to near virtualized servers, a strategy called close-coupled cooling. The idea behind the close-coupled cooling is to have more control over the supply of cold air and warm air capture. You can employ the right cooling required for a particular rack at the time that it takes.
4. Fill voids
When virtualized servers are removed from a rack, you can create problems with the cooling and energy efficiency. Fill the empty spaces within racks, cabling, panels etc is vital in this scenario.
5. Adapt approach
Virtualization facilitates the creation of zones. So you can have a highly used area within the data center, which will allow literally close all other areas of the environment. In turn, the appropriate power and cooling capabilities can be located precisely where they are needed. Zones organized according to applications can also allow better load balancing data center, server efficiency and redundancy requirements. For example, back office applications can be setup in an area while mission-critical systems in another. Instead of delivering the same amount of redundant power and cooling throughout the mission critical zone can be targeted to the highest levels of redundancy and cooling. The opportunity to save energy in the data center requires looking at more variables. Efficiency load and redundancy work together to optimize energy consumption.
6. Save with air coming out
Besides taking care of the internal air, which comes from outside should also be taken into consideration. Data centers are often maintained by increasing the temperature according to the needs of the equipment. Meanwhile, data centers can save to rely on outside air for cooling. Increasing the internal temperature of five degrees will not compromise the performance of the equipment IT, and will allow the data center to enjoy the days of cold outside to keep the proper temperature.’s economy generated using this method can be seen quickly. It is possible to record energy savings of 50% or more every hour that the company use the economy mode. Optimize energy use across the data center can be a time consuming exercise, but following these relatively simple steps you can register major economies.