English | EN 
The cloud is becoming increasingly popular in the workplace, not just because of the cost-cutting opportunities but also because of its productivity and efficiency benefits. One great example is Kubernetes which has rapidly become one of the most popular technologies for managing workloads and streamlining IT infrastructure. A good software service like Kubernetes can be deployed on a public cloud, and many companies have made the move to public clouds to get better performance and lower costs. However, some enterprises are concerned about security since they do not control the infrastructure that they are running their workloads on. Managing workloads using a cloud is a natural progression.
Data Centers and Carbon Emissions
Data centers are projected to account for 10% of global power consumption by 2022, resulting in 200 million tons of CO2 emissions. All data centers, however, will not result in the same amount of emissions due to their size and individual approach.
For example, companies who use public clouds for their IT infrastructure have lower CO2 emissions than those who use private data centers. A study showed that organizations which opted to run their workloads on the cloud reduced their carbon emissions by an impressive 88%.
The reasons cited in that article are higher server utilization numbers, a lower power usage effectiveness (PUE) and a more energy-efficient mix in cloud data centers. Some analysts argue that tech can in fact be bad for the environment, as easy provisioning of cloud services leads to more consumption and a correspondingly large carbon footprint. .However, the article also makes a point to mention that tech companies are working on reducing their environmental impact. The article cites how Microsoft drives various initiatives focused on energy efficiency and renewable energy.
Can Kubernetes Improve on the Cloud?
Is Kubernetes powered by the cloud better for the environment than an on-premises data center? More specifically, can it help companies reduce their carbon emissions? To find out, let’s dig into how carbon emissions work in both cloud and on-premises environments.
Carbon emissions in the context of both on-premises and cloud data centers are a function of four variables: Power Usage Effectiveness (PUE), server utilization, carbon emissions intensity of the electricity grid they are connected to and the overall power consumption. Let’s take a closer look at each of these variables.
Server Utilization
Server utilization numbers can be really tricky to come by and can vary greatly depending on which resources are being utilized. One study mentions that the best practice for using cloud data centers is around 70%, and the worst-case scenario is under 7%. In contrast, cloud environments have a utilization rate that typically varies between 5%-25%
Server utilization varies even more across individual customer environments in the cloud. Take Nordstrom for example which ran 1000s of VMs on AWS with an average CPU utilization of 4%.
Power Usage Effectiveness
PUE is the measure of how much power it takes to run a data center versus how much the actual servers need. PUE can be high for inefficient data centers, but there are some ways to lower that number.
Lots of companies outsource their data storage to the public cloud because it allows them to focus on selling and developing other aspects of their business. Lots of industries use public cloud facilities due to the benefits they provide. Having your own data center may not be as efficient as some people think it is
Carbon Emissions Intensity
When considering the carbon intensity of different power sources, one should bear in mind that most data centers are plugged into local power grids. Therefore the main factors influencing this will be the fuel mix used in generating each region’s electricity supply. However, there are a few exceptions to this rule. For example, when cloud providers and some data centers source power from renewable energy systems set up specifically for that purpose.
Containerization is the latest trend, with Docker being the most popular option in data centers – which are a form of virtualization. Containerization can help with resource management and allocation, but it can also end up increasing carbon emissions.
How Does Kubernetes Help?
The introduction of Kubernetes into cloud environments has implications for two of the three variables highlighted in the previous section: server utilization and emissions intensity. We will examine both in the following paragraphs.
The National Resources Defense Council believes server utilization to be the single most important factor to determine efficiency and thus carbon footprint. Higher utilization means less resource wastage, fewer machines, reduced infrastructure footprint and less power required to run it. This, in turn, leads to a reduction in the carbon emissions associated with operating infrastructure.
Kubernetes is becoming a popular option for enterprises, who often adopt it due to its efficiency and speed. In fact, this is one of the top features that attracts them to Kubernetes.
Clearer explanation of the change could have something to do with Amazon Web Services and a retailer, which had better results after they started using Kubernetes. They ran thousands of virtual machines on AWS before switching to Kubernetes but averaged only 8% CPU utilization. This changed to an average of 70% post-kubernetes.
The high utilization means that Nordstrom can operate the same workloads with only 1/10th of the VMs they needed pre-Kubernetes and allows them to scale down their infrastructure. This does wonders for the carbon footprint of Nordstrom itself, which can now report a 90% reduction in the carbon emissions associated with their use of AWS VMs.
Another variable associated with carbon emissions that Kubernetes can potentially impact is emissions intensity. Since Kubernetes is inherently portable, it allows much greater control over decisions about workload placement.
In the earlier post, I mentioned that forest management practices and other large-scale effects can affect carbon emissions intensity. But the small scale impact is most visible on a regional level. ITDMs can use open emissions APIs to evaluate regions based on their emissions intensity and make workload placement decisions accordingly
The Low-CO2 K8 Scheduler is worth looking into for this purpose, as it makes scaling decisions based on CO2 emissions intensity in different regions. There are a lot of reasons to choose a carbon-neutral cloud provider. For example, when you scale up your workloads in the designated area, there won’t be any more risk of contributing to climate change and you will also get access to better performance servers.
Conclusion
With the world’s data centers expected to consume 3 percent of electricity in 2022, that translates to 100 million tons of carbon dioxide emissions. Cloud data centers operated by hyperscale IaaS providers score much lower on the emissions scale, which has a significant impact on the climate. Azure datacenter in North Carolina uses 150 megawatt, but less than 18 megawatt is generated by coal-fired power facilities and the rest comes from natural gas fueled power. plants. At the same size, Azure in North Carolina spends more than $234 million annually on purchased electricity to power its cloud infrastructure, which has a negative environmental impact of about -0.05 tons per machine hour and an annual potential for carbon dioxide emissions mitigation of roughly 120,000 metric tons. Beneath the hood: How Microsoft Azure’s cloud
However, there is certainly room for a lot of improvement.
In this blog post, we reviewed the impact that Kubernetes can have on two of the biggest factors in determining carbon emissions in a data center – server utilization and emission intensity. Kubernetes significantly improves server utilization leading to a smaller infrastructure footprint and a corresponding reduction in carbon emissions. Organizations are able to choose where their servers are located for greater control over workload placement. These servers can be placed in areas with lower emissions or a smaller carbon footprint. There are many other factors that have an impact on CO2 emissions. Kubernetes can also significantly affect how much energy is consumed by servers. Server-less computing is popular because it means data centers require less processing power. The result is enterprise’s carbon emissions are smaller too, Kubernetes offers some control over CO2 emission levels.
