Kubernetes: The Cloud-Native Hydra
Cloud-native moves fast, doesn’t it?
But each KubeCon is an opportunity to pause in this frenetic race and take stock of how far the community has come.
This year, as we look ahead to our gathering in Paris, it’s inarguable that Kubernetes is not just evolving; it’s mutating.
Our research found that 56% of businesses have more than 10 Kubernetes clusters, and 69% run Kubernetes in multiple clouds or other environments—with 80% expecting their scale to grow.
Kubernetes is spreading into all kinds of different use cases, each with very different environments and challenges. It’s a testament to the versatility of Kubernetes as the “de facto OS” for the cloud-native era that it can adapt to all these environments. You could even see it as survival of the fittest in action.
SaaS, PaaS … KaaS?
Inside the enterprise, in the world of platform engineering teams, we see companies using Kubernetes as a driver for standardization. They see, at last, one technology, enhanced by universal translators like CNCF’s Cluster API, to bridge their multiple public and data centers, bringing a shared ‘language’ to a world of silos.
More importantly, they see Kubernetes, when built right, as the engine for a new kind of experience. You could call it ‘clusters-as-a-service’ or Kubernetes-as-a-service (KaaS), at incredible scale: A streamlined, even self-service experience meeting the needs of dozens of different business units and thousands of developers who just want to deploy their code in a safe, easy way.
For these enterprise platform teams (and really, this is what’s driving the trend toward ‘platform engineering’), the challenge is how to achieve, on the one hand, the velocity and choice that devs want and, on the other, the control and consistency that the business needs.
It’s far from easy. With scale comes complexity, and with complexity? Well. A full 75% suffer from interoperability issues that affect the running of their clusters, up from 66% in 2022. And 40% say they lack the skills and headcount to manage Kubernetes.
Solve this equation? It’s an intoxicating opportunity: The key to boring old IT evolving from a cost center to a driver of measurable innovation and business value.
Kubernetes InsideⓇ
Elsewhere, we see ISVs of all shapes and sizes using Kubernetes and containers as a delivery vehicle for their customer-facing software, sometimes built into hardware as a unified deployable stack, sometimes hosted as SaaS for different tenants to access, equally often deployed locally in client sites—but always multitenant.
They’re looking to the portability of containers and the resilience and scalability of Kubernetes to deliver a quality customer experience in a world that has, at last, defiantly left virtual machines behind.
But how do you make sure that customers have the right experience, no matter what their environment looks like? How do you keep their instance up and performing and their data secure? And, most importantly, how do you do that without burning all your margin along the way?
Bare Metal is Back
Many organizations are starting to wonder if Kubernetes is the tool to unlock a return to bare metal, promising an end to hypervisor overhead (both performance and financial) and just maybe a new home for their orphaned VM workloads for hybrid app development, too.
We can’t think what could have triggered this … let us know if anything has happened, will you?
Whatever unspoken event is the cause, there’s fuel on the fire. Companies are earnestly looking at bare metal. They’re looking at projects like KubeVirt. Now is the time to modernize and rethink 20 years of legacy before the burning platform sinks for good.
And when they’re looking at this shift? It’s time to avoid making the same mistakes again. Lock-in and proprietary solutions won’t serve us in this new era. Choice, open standards and portability are the new requirements.
From Cloud to Edge
For other businesses, Kubernetes is the gateway to making edge computing happen at scale, with freedom from proprietary and manual stacks and from manual processes.
They’re putting K8s and containers on the ‘fat edge’ in micro data centers and even on small form-factor devices under counters and in backroom cupboards across the built environment.
Here, what keeps them up at night is the cost (of deployment, field engineering, downtime) and that ever-present fear about security.
Open source projects, community organizations like LF Edge, government bodies and bold hardware and software vendors are tackling these challenges with gusto. They know their efforts matter, unlocking the economics of edge scale.
And already, we’re seeing triumphs. Companies like agritech startup Tevel or dental equipment manufacturer Dentsply have cracked the code. They are seeing results others can only dream of.
All Things to All People
These are different worlds, with different challenges, different stacks, different teams, different hardware, different scales. Wherever you look, Kubernetes is there; it’s the one common factor. This is the real indicator of Kubernetes maturity: It’s evolved into a chameleon. And an unkillable one.
Kubernetes has many different faces; that’s its strength. Those that focus on the distribution or who get opinionated about the stack are missing the point: Whether you use an edge distro like K3s or roll your own from pure upstream, it doesn’t matter. The more the merrier, and for each situation, there’s the perfect distro out there—just as there is the perfect stack of service mesh, observability, storage and many other layers. Already, our research found that 39% of businesses have more than 10 cloud-native integrations in their Kubernetes stacks, and 83% use multiple Kubernetes distributions.
We as a community need to focus on this jungle of cloud-native innovation as the strength it is, rather than an enervating force of complexity. Let’s embrace the choice, embrace the mutation. Hail hydra!
Explore the Latest Innovations
We’ll be at KubeCon in booth J24 and at Edge Day on Tuesday, March 19, before the main event starts. Join us to discover our latest innovations and cutting-edge open source technologies, or come and watch our sessions on Friday, March 22, where we’ll be introducing a new two-node HA edge architecture. Our customer, Dentsply, will be sharing how it built a scalable edge infrastructure on Kubernetes. Check out all our activities right here: https://info.spectrocloud.com/kubecon-europe-2024-paris.
To hear more about cloud-native topics, join the Cloud Native Computing Foundation, Techstrong Group and the entire cloud-native community in Paris, France at KubeCon+CloudNativeCon EU 2024 – March 19-22, 2024.
