Designing a modern, sustainable IT infrastructure in AWS
How migrating and modernising your applications with AWS can help you hit your sustainability targets, as well as unlocking a range of business benefits.
It’s time to stop talking about sustainability and start doing it. Perhaps that’s a bold opening for a blog that focuses on making your IT environment more sustainable, but the point is that we need to move beyond talking about what’s possible, and start planning how to do it. In this series of articles, we’ll be showing you how you can make great strides in making your applications and infrastructure more sustainable by modernising them in AWS. To help make what we’re talking about feel even more real, we’ll be sharing with you a fictionalised-yet-completely-plausible example of a company migrating their applications into the cloud, working with Claranet and AWS.
The link between modernisation and sustainability
When you think about how to make your IT environment more sustainable, you probably start by thinking about how you power your technology, keep it cool, and so forth – in essence, pulling levers at the hardware level that help you use less power or use that power more efficiently. There are other levers you can pull, however, at the software and platform level of the stack, which further optimise your power consumption. Modern, cloud-native applications and platforms are designed to use resources only when needed – even for application infrastructure that is traditionally always on, such as disaster recovery – while still giving the same quality of experience to the end user. Because they use resources more efficiently, the power you use and hardware you need – and therefore your carbon footprint – are further reduced.
Why AWS?
If you’re already running a private cloud, you could attempt to modernise your estate yourself. However, public cloud providers like AWS can help you both modernise and make your estate more sustainable quicker, more effectively – and most likely cheaper – than you could achieve on your own. This is because:
- Migrating to AWS means adopting an infrastructure-as-a-service model. That means that many of the modernisations you might plan (and budget) for your private datacentre have already been implemented by AWS – you get all the benefit, with none of the investment you would have to sink into a private cloud to get it performing at the same level.
- AWS can scale its operations in ways you couldn’t even begin to do. Its size means it can access scales of economy when it comes to cooling, real estate, and power consumption that enables AWS to run its data centres incredibly efficiently.
- Because AWS exists to provide cloud services to its customers, it’s focused on innovating to reduce its energy consumption and so be more sustainable. For example, it’s designed a new chip – the Graviton3, that can reduce power consumption by up to 60%, and created a new cooling medium that reduces cooling power consumption by 20%.
AWS has been making great strides recently to ensure that it runs its data centres as sustainably as possible. It has committed to net-zero carbon emissions by 2040, to power its operations with 100% renewable energy by 2025. A recent project saw AWS become the world’s largest corporate buyer of renewable energy, and it also announced a new data centre in New Zealand that will be powered by 100% renewable energy.
Starting your journey to a sustainable infrastructure in the cloud
Step 1: work out your current state
The first step in any sustainability transformation is to understand the current carbon footprint of your IT infrastructure. There are multiple angles to consider when calculating this:
- Exactly how much hardware is included in your estate?
- Do you know where all your servers are? Ensure you look thoroughly across your organisation to uncover any shadow IT.
- How much power are you using to keep your estate running? This includes running the servers in your data centre, but also the power used to cool your equipment, light your server rooms, and so forth.
- What proportion of that power is generated from sustainable sources?
- How much renewable electricity do you buy?
- If you have a dedicated facility housing your data centre, what’s the embodied carbon level in the building?
- Embodied carbon is materials which generated emissions during manufacture which could be reused, such as glass, steel, and brick.
- If you have dedicated staff running your data centre, do they drive to work? What emissions do they generate?
Of course, alongside gathering this information it is also prudent to gather information about the costs of running your IT estate. In an ideal scenario moving to the cloud and making your estate more sustainable will also deliver cost savings, which will be vital when it comes to justifying the migration to the business.
Step 2: Build the plan
Once you have a handle on your current state, you can start to think about your transition. Start by prioritising your applications for migration. You may want to prioritise by business criticality, or by power usage to quickly reduce your carbon footprint.
It’s important to start thinking about what your migration will actually look like for each application – because there are different options. For instance, you might just move an application as it is over to AWS, known as rehosting. Or, you might choose to discard your current application and pay for a new application, which we refer to as repurchasing. Or, you might rebuild your application to take advantage of cloud-native features, which we call refactoring. In fact, at Claranet we have seven different options for applications when planning your migration – our “7 Rs”. We’ll cover them in more detail in our next blog, but for now it’s enough to say that each of these options has different consequences for how sustainable your new IT landscape will be.
Alongside this, start to look at the skills in your organisation. What skills might you need to bring in to make the migration a success – for instance, do you have anyone with AWS experience already who can help you migrate and then run your new infrastructure? You’ll ideally need experience in some specific technologies including Apps2Container (which we’ll cover in more detail in our next post), Elastic Compute Cloud (EC2), Elastic Container Service, and Fargate. If you can’t find these in-house, you’ll need to invest in training or look to a trusted partner who can bring these skills to your migration.
Finally, at this stage you’ll want to project what improvements you expect to see from your migration – both in terms of carbon footprint and in terms of performance.
This information, combined with the picture of your current state, will give you the bones of your business case. That document will give your migration project a clear plan and set of goals, and also help stakeholders understand the benefits you stand to gain, making it easy for them to buy in to the plan.
How might this look in real life?
This all sounds great, I hear you think – but talk is cheap. To help it feel less like a pie in the sky idea, I’d like to introduce you to Charlie Loud Migrations Ltd. C. Loud Migrations is a (technically hypothetical but very plausible) business that specialises in tracking and reporting on animal migration patterns for conservation and development purposes. Across this series of blogs, we’ll be following C. Loud Migrations as it makes its own journey from a private data centre to AWS.
Charlie Loud, CTO and founder of C. Loud Migrations, knows that the company needs to create a more sustainable infrastructure. On the one hand, sustainability is very important to the conservation groups and public sector organisations that make up the bulk of its customers. On the other hand, as CLM has scaled to track more and more animals, leveraging IoT technology and other innovations, its IT estate has sprawled and Charlie knows the team could be running things more efficiently from an operational perspective as well as a green one.
CLM decides early on that it will need support with this endeavour. So, it appoints Claranet as its cloud modernisation and migration partner – and quickly chooses AWS as its cloud provider.
The first step in the process is to map out the existing architecture. Together, Claranet and CLM uncover a total of 75 separate applications running across the business. That includes 25 SaaS applications that Charlie and the IT team weren’t actually aware of, and the business’ core application - a custom-made piece of software to help them manage all the data inputs from the tracking devices they attach to animals and analyse that data. The application runs on a legacy server, and due to its complexity has never been updated. The team reviews the business’ power usage, and discover that it is not on a 100% renewable energy contract from a reputable supplier.
With a complete view of its IT infrastructure, C. Loud Migrations can calculate the true carbon footprint of its IT estate – and works out that it can reduce its energy usage by around 85% by moving to AWS and modernising its applications.
CLM recognises that it needs to create a steering group of cloud experts to help it migrate and modernise, and then get the most out of its AWS technology. Claranet and CLM have already reviewed the skills in the business and discovered a group of individuals with previous AWS experience. However, Charlie and the team know that they have nobody who is familiar with containerisation, which they know will be vital to creating serverless applications further down the road. Claranet agrees to help C. Loud Migrations staff its Cloud Centre of Excellence with Claranet experts who can fill those skills gaps, so that from the very beginning of the project Charlie and their team are making decisions based on best practices and the latest technology innovations.
Working with the CCOE, Charlie and their team prioritise which applications to migrate first. They choose their door entry system as a proof-of-concept to demonstrate that the process is valid, and (assuming all goes well) they then choose to prioritise the custom application at the core of the business, for two reasons: first, so that they can then migrate the various apps that depend on the core application more successfully, and second, so that they can quickly turn off the servers on which it currently runs, as those servers are some of the least efficient pieces of technology in CLM’s estate. At the end of this work, the team creates a formal business case to present to C. Loud Migrations’ board of directors. The business case states that they hope to reduce power consumption by 85%, increase their use of electricity generated from renewable sources, and reduce the heavy lifting that the team has to do in order to manage their estate. The business case, we’re pleased to say, is accepted and the project is a go.
What’s next?
In our second article, we’ll cover the actual migration itself, and what options you have with your applications as you move them into AWS.
Find out more about AWS solutions tailored for your organisation.