While still in the middle of mastering their digital transformation journey, Airports are increasingly facing the need to also meet sustainability goals. With data centre operations being the backbone of digitisation, the challenge is to find the right balance between these two initiatives.
Digitisation and sustainability are closely linked. On the one hand, the transformative power of digital technologies makes it possible to achieve sustainability goals in a way that were previously unthinkable. On the other hand, digital infrastructure consumes valuable resources both for its production as well as in operation. For this reason, data centre managers need to review the efficiency of their infrastructure to avoid a boomerang effect, where supposed benefits through increased use of digitisation can lead to an increase in resource consumption.
In addition to the challenge to balance digitisation and sustainability goals, the cost of energy is increasingly becoming a hard business factor. The times when enterprises could obtain their electricity comparatively cheap are over. With an increase of 58% for enterprise data centre owners and 64% for colocation providers, Uptime Institute’s Data Centre and IT Spending Survey 2022 showed power to be driving the greatest unit cost increases – the result of high fuel prices, thetransition to renewable energy, imbalances in grid supply, and the war in Ukraine. The UK and the EU have been most heavily affected by these increases. Despite all the recent governmental countermeasures, it is doubtful electricity prices will reach pre-corona levels again, they are likely to remain well above the average levels of the past two decades.
The question of how to improve energy efficiency comes up in almost every Airport and Aviation IT project today. According to IDC, by 2026, ESG performance will be viewed as a top 3 decision factor for IT equipment purchases and over 50% of RFPs will include metrics regarding carbon emissions, material use, and labour conditions1.
This white paper outlines how integrated systems help IT operations to introduce a more efficient infrastructure – regardless of whether deployed in the core data centre or at the edge. Besides the great impact of introducing the latest energy-efficient server systems, it shows the benefits of modern architectural concepts such as hyper converged infrastructures or hybrid clouds and also explains the value of adopting an “as a service” IT consumption model.
1 IDC FutureScape: WW Sustainability/ESG 2023 Predictions | IDC Blog
Replacing older systems
Servers are the number one energy consumers in IT. A study by the Borderstep Institute commissioned by Bitkom 20222 in Germany found that servers accounted for almost 40% of the overall data centre energy consumption. Thus, they offer the greatest potential for energy-efficiency gains and footprint compression.
For this reason, it is worth considering a tech refresh as modern servers are not only more powerful than older systems: they are also far more efficient. A single server of the current generation can take over the tasks of several legacy systems simultaneously, which reduces your data centre footprint. And because modern systems also include more efficient hardware components and optimised power-saving functions, a tech refresh leads to significant savings in power consumption in your data centre and is therefore crucial for sustainable data centre operations.
Selecting the right data centre architecture
When undertaking a tech refresh, it is often quite sensible to also review the effectiveness of the underlying data centre architecture. For decades, a traditional 3-tier architecture comprising server, external storage, and networking gear has been a proven foundation to run data centre workloads. Converged infrastructure approaches powered by server virtualization technology brought a great leap forward in terms of sustainability. However, cutting edge architectural approaches like hyper-converged infrastructures promise to further improve efficiency gains.
Not so long ago, it was common in data centres at Airports around the world to run applications directly on physical servers that often sat mostly idle. Within a decade, the use of software-defined compute, storage, and networking approaches have become common industry practice. For example, server virtualization improves resource utilisation, accelerates resource provisioning,enables workload mobility and makes it possible to implement affordable high-availability solutions. But more importantly, it also reduces server sprawl and data centre footprint, which brings major improvements when it comes to reducing energy consumption and cooling requirements.
Nowadays, in “converged infrastructure approaches,” server virtualization is a key element of the solution stack. A converged infrastructure is a pre-packaged bundle of 3-tier systems, including servers, storage, networking, and management software. Companies usually purchase these systems from one company, instead of buying the hardware and software components separately from different suppliers. The beauty of a converged infrastructure system is that it typically comes pre-configured and pre-tested, making it easier and faster to deploy when building out a data centre.
As shown above, a classic converged data centre infrastructure consists of servers, external storage systems, network components, and software. While using server virtualization software optimises the compute part, there are still all the individual components of the storage infrastructure (storage arrays and network) that consume energy for power and cooling.
This is not the case with hyper-converged infrastructures: by combining software-defined compute and storage technology, hyper-converged infrastructures tightly integrate all compute and storage resources in a cluster of commodity x86 server nodes, making a dedicated physical storage area network (SAN) with its management superfluous. Instead, storage is spread across the local disks of the server nodes. As there is no external storage involved, the data centre footprint is reduced, as are energy consumption and cooling requirements.
Adopting an “as a Service” IT consumption model
Another opportunity to build sustainability considerations into your digital agenda lies in the way you procure and consume IT. The market is currently seeing a significant shift toward the adoption of as-a-service IT consumption models for on-premises IT infrastructure. IDC predicts that by 2026, 65% of tech buyers will prioritise as-a-service consumption models for infrastructure purchases to help restrain IT spending growth and fill ITOps talent gaps2. Alongside the flexibility and cost control of as-a-service models, their environmental benefits that help companies achieve their sustainability goals are often underestimated.
Below are some reasons to shift to an “as-a-service” IT consumption model from a sustainability point of view:
1. Eliminate overprovisioning
Many enterprises struggle with on-premises capacity planning of their IT resources resulting in IT production environments statically overprovisioned by 30% to support growth3. These inefficiencies not only result in redundant CapEx and OpEx costs, but also negatively impact power, space, and cooling requirements. By adopting an “as-a-service” IT consumption model, enterprises can overcome the challenges of overprovisioning. Based on an in-depth review of your needs, you should receive appropriately dimensioned IT capacity from the start including built-in buffer capacity
calculated according to your growth plans. In addition to significant CapEx savings, “as a service” IT consumption helps companies avoid the waste of resources and costs associated with both unused equipment and unnecessary maintenance contracts and licenses.
2. Increase resource utilisation
Low resource utilisation is another common problem IT organisations are facing. Industry-wide utilisation levels are still estimated to be around just 25%. This means that three-quarters of IT resources are not being used effectively, leading to additional wasted costs and resources. These issues can be addressed by assigning a Customer Success Manager (CSM) to each customer. The CSM is responsible for managing the ongoing relationship with the customer. Being a single
point of contact for any issues or escalations, this individual also works closely with the customer to review current and future resource requirements and make recommendations on how to optimise resource usage. Customers can also be provided with access to a “as a service” Portal. Besides important information on billing, support, contacts, and order updates, this portal also gives an overview on current consumption data.
3. Accelerate access to latest energy-efficient equipment
Consuming IT on an as-a-service basis also takes the burden of IT lifecycle management off your hands while ensuring you get accelerated access to the latest technologies, which typically consume less power than the previous generation. As outlined in the sections above, a tech refresh reduces the performance lags and inefficiencies in power and space consumption that often accompany aging data centre technology.
4. Extend life of retired systems
In order to manufacture IT equipment, you need enormous amounts of raw materials and energy which pollute the climate with carbon emissions. This is why we should take every effort to extend the product life cycle to conserve natural resources and reduce emissions. Depending on the customer contract conditions, your provider retains ownership of the IT equipment upon termination of a as a service contract. Choosing a partner who is working with regional 3rd parties who are specialised in picking up used IT devices, which will then be refurbished and remarketed for reuse after a certified data deletion is an additional way to lessen the environmental impact. Devices that are no longer marketable will be handed over to a certified recycling company.
2IDC FutureScape: Top 10 Predictions for the Future of Digital Infrastructure
37 Reasons Cloud Budgets Don’t Stay in Check | Gartner
Taking advantage of hybrid cloud
It is common knowledge that the carbon efficiency of the data centres of the hyper-scalers is significantly better than most company-owned data centres. Although the average annual power usage effectiveness (PUE) of data centres has seen a massive decline from 2.5 in 2007 to 1.55 in 2022, new data centre builds of the hyper-scalers routinely outperform the average, achieving PUEs of 1.3 and below using facility designs and more advanced equipment that are optimised for lower energy use4. More than two-thirds of this advantage is attributable to the combination of a more energy-efficient server population and much higher server utilisation. Moreover, hyper-scalers usually have comprehensive efficiency programs in place that touch every facet of the facility. They can afford to deploy a dedicated team of engineers which can entirely focus on energy efficiency – something which is often not possible in company-owned data centres due to budget and operating expenses constraints.
For this reason, it also makes sense from a sustainability point of view to review your application landscape and introduce a hybrid cloud approach by moving selected workloads to the cloud. Key to this is having the option to deploy platforms on VMware, Microsoft, and Nutanix technology to integrate your on-premises deployments with cloud systems from the major hyper-scalers like Amazon, Microsoft, and Google.
4Uptime Institute Global Data Centre Survey 2022
Summary
The operation of IT whether deployed in the core data centre or at the edge requires significant energy and resources. This though does not mean that many sustainability goals can still be achieved. The digital applications running in data centres do not only allow savings in energy and greenhouse gases in almost all areas of life and work. Moreover, they support a resource-saving industrialisation and drive innovation, thus making them a critical part of a sustainable and resilient infrastructure. In light of their growing importance, it is necessary to make the operation of data centres as energy efficient and climate friendly as possible.
In the future, more stringent sustainability regulation and reporting requirements will force IT to deliver improved performance in energy efficiency. Ongoing price pressures – unlikely to abate, even in the long term – are a further driver. Compared to other measures, IT still has considerable scope for delivering improved energy efficiency. In fact, every watt saved by IT reduces pressures elsewhere.
By introducing state-of-the-art architectural approaches supported by highly efficient hardware along with flexible IT consumption models, Integrated Systems help you balance digitization and sustainability goals while keeping your energy costs under control.
For more information please contact: ben.huntley@fujitsu.com
