One of my favourite authors, Douglas Adams, once said that “we are stuck with technology when what we really want is just stuff that works.” Whilst Adams is right about a lot of things, he got this one wrong – at least when it comes to infrastructure. As our Infra Masters 2025 event demonstrated, infrastructure is the technology that makes everything work – from managing a satellite in outer space, to, say, livestreaming an event.
Held at Canonical’s London office on March 31st, Infra Masters 2025 brought together operations leaders and architects to explain how to build infrastructure that transforms industries.
If you didn’t attend the event, don’t worry – and, naturally, “DON’T PANIC.” You’ve come to the right place to find out what you might have missed. You can watch the full talks on YouTube, or read this article for an overview of everything that took place, from key insights from ESA on modernizing infrastructure to BT’s network cloud transformation.
So, without further ado, what can we learn from Infra Masters 2025?
1. Reconsider your vendor relations to speed up innovation
Modernizing infrastructure isn’t just about choosing the right software. It’s also about who you choose to work with, and how you work with them. According to BT, it was a fundamental shift from vendor-consumer to partnership with Canonical that invigorated their efforts to modernize their infrastructure.
As their representatives disclosed in the first talk of the day, BT have worked with Canonical since 2019 on their ongoing infrastructure transformation. Despite the challenges posed by the transition, the collaboration between BT and Canonical has been marked by open communication, shared goals, and regular training sessions to upskill engineers. The secret to a successful partnership?
“It’s the collaborative approach, working together working on shared goals.”
– Curtis Haslam, Network Cloud Senior Manager, BT Group
BT was keen to emphasize that transparent collaboration requires willingness to offer and accept constructive feedback. Haslam explains, “we’re very honest”, because “acknowledging mistakes from both sides” is the best way to fix errors. As Douglas Adams puts it, a learning experience is “one of those things that says, ‘You know that thing you just did? Don’t do that.’” However, this only becomes possible through the mutual trust that grows from a long-term partnership.
2. Double your output with Kubernetes
The European Space Agency (ESA) plans to double the number of missions they want to run by 2030 – no easy feat. With critical projects covering everything from searching for habitable worlds, to clearing the 130 million pieces of orbiting debris that threaten satellites, each mission had its own individual compute and infrastructure needs, making their goal a particularly ambitious one. For organizations interested in how to increase output by modernizing their infrastructure, ESA’s presentation may provide some helpful tips.
As Michael Hawkshaw, ESA Mission Operations Infrastructure IT Service Manager at ESOC (European Space Operations Centre) explains, with Canonical Kubernetes ESA has been able to automate the deployment of both infrastructure and “all the software needed for those missions as well.” Canonical Kubernetes readily plugs into Ceph and PostgreSQL for instance, which are part of ESA’s stack. These automations have, naturally, freed the team to work on other mission-critical tasks. Likewise, by increasing availability and reducing “wasted space” on database servers, Canonical Kubernetes have helped ESA to support more missions.
Want to try Kubernetes but don’t think you have the capacity? ESA was in the same position. Michael acknowledges that Kubernetes is fast-moving and has a steep learning curve. However, with Canonical managing the systems for them, on call “24/7, 7 days a week”, ESA can always “get the support” they need, “with active monitoring” of their setup.
For more details on the ESA’s infrastructure modernization, check out the case study.
3. Open source software empowers efficient infrastructure
Given 70% of organizations find that “open source is extremely important to run mission-critical workloads” as our recent report in partnership with the International Data Corporation (IDC) and Google Cloud highlights, it was unsurprising that the universal, underlying theme behind the event was the crucial role that open source software has to play in infrastructure modernization.
For one thing, open source software gives organizations the flexibility to scale and adapt to shifting requirements. To return to our initial Adams quote, while it may be true that with proprietary solutions, you can become “stuck with technology”, you’re never stuck with open source.
For BT, open source was critical to building the Network Cloud, the infrastructure project that helped them to achieve their goal of bringing 5G to the UK. The Network Cloud replaced a variety of disparate, proprietary vertically-integrated stacks. These stacks each required individual management, oversight, compliance, authentication, and deployment, making them time- and cost-intensive. The challenge was to replace these with infrastructure that was highly dependable and automated, providing consistent high performance .
The answer was consolidating their infrastructure into a single, trusted open source stack – including MAAS for bare-metal provisioning, Ceph for storage, LXD for container management, and Juju for automation. James Cawte, BT Group’s Cloud Network Principal Engineer, noted that by streamlining their operations in this way, the app developers were free to ‘focus solely on their application development’, rather than trying to make the infrastructure work – which allowed BT to streamline its operations. For more details on BT’s partnership with Canonical, explore the case study.
As Canonical’s Thibaut Rouffineau noted, open source software helps organizations to scale resources whilst keeping costs down. BT and the ESA demonstrate that moving away from proprietary software reduces expensive licences and contracts, whilst enabling companies to optimize their infrastructure – rather than getting stuck with technology, it just works.
4. The future is in the clouds, on the edge, and managed by Kubernetes
…When it comes to infrastructure, that is.
Moving forward, BT’s focus is on enhancing edge computing capabilities for better 5G performance, optimizing infrastructure for containerized applications, and integrating serverless computing to improve developer workflows.
This shift to edge computing seems likely to become increasingly common as organizations choose to move away from the public cloud and data centers, and distribute their infrastructure across edge devices. Kubernetes, and the automations it makes possible, will form a key part of managing this infrastructure, offering a new approach to how we think about the future of technology.
Meanwhile, ESA’s plan to double the number of satellites it currently flies by 2030 relies on Kubernetes and cloud-native computing. AI tooling managed by Canonical Kubernetes has increased the amount of data that can be stored and retrieved, whilst Ceph and PostgreSQL support these cloud-based workloads. As space exploration continues to evolve, ESA can more easily scale its workloads thanks to these tools.
Infra Masters: the conclusion
So, to take a final leaf from Adams, and “summarise the summary of the summary,” what can we learn from Infra Masters?
The relationship between vendor and client is critical, and moving towards a more collaborative partnership can improve innovation, efficiency, and in-house skillsets. Equally, the support provided by a company like Canonical to aid migration efforts, for example, can help to take the pressure off in-house engineers to avoid disrupting workflows.
The automations enabled by Canonical’s infrastructure portfolio improve efficiency, reduce costs and take the pressure off managing infrastructure. Choosing a managed solution can help organizations to get these benefits, without worrying about capacity or skill shortages.
Open source software is an increasingly important part of the stack for many companies, providing cost savings, the opportunity to scale, and even create a unified platform and integrate infrastructure seamlessly across different environments.
As the company behind Ubuntu, Canonical’s software is widely used, trusted, and provides a great option for organizations looking to explore open source options for their infrastructure – whether it’s to lower costs, gain architecture freedom or cloudify their data center. And that’s all, folks. So long, and thanks for all the fish – and by fish, I mean “humoring my … creative references.” See you at the next Infra Masters!