Smooth Sailing at the Edge: How to Migrate Legacy VMs to Containers with Avassa

Over the past decade, containerization has revolutionized the way applications are deployed and managed. While the shift toward containers has been rapid in cloud environments, edge computing has traditionally moved at a slower pace. However, the tide is changing. A combination of technological advancements and evolving business needs is driving a shift toward a container-first approach in edge computing. In this article, we explore the key forces behind this transformation and how containers are shaping the future of edge application management.

Fact is, we see a number of drivers toward container first approach.

In the cloud, we already see containerization taking over from a VM approach. The benefit of the shift is clear—no more operating systems to manage, as that responsibility falls to the cloud provider. Instead, developers and businesses can focus on application development, while the cloud providers will give you all the tooling you need.

On the other extreme, IoT and Industrial IoT, have traditionally been built using classic embedded tooling. Build an ultra-light OS, adding only the necessary components, often written in C, to maximize efficiency and minimize resource consumption. This was a necessity when hardware constraints were severe.

Nowadays, you get a lot more CPU and memory per dollar spent, we see an increasing drive to use e.g. IoT gateways as more general compute platforms. The customers of the IoT hardware often want to run some applications of their own, in addition to the workloads that the IoT vendor provides. This is the perfect scenario for containers.

Given these two drivers, the case for a container-first approach at the edge is compelling. Containers provide a lightweight, efficient, and highly portable way to package and deploy applications, whether they are running in a data center or on an IoT gateway in the field, or in-between at the on-site edge. With modern container orchestration tools, edge deployments can benefit from the same automation and lifecycle management capabilities that have made cloud-native applications so successful.

But then…

Of course, reality is never quite that simple. While we may wish for a purely containerized future, the fact remains that some applications still require virtual machines to function. This will certainly go away over time, but when is anyones’ guess.

Let’s just stop for a quick sanity check, and look at the main differences between a container and a virtual machine.

Recognizing the need for flexibility, we decided to ensure that our platform could support both containers and VMs in a unified deployment model. We found that containers provide an elegant solution for distributing VM images as well. This hybrid approach allows users to package VM workloads inside a container, simplifying deployment while maintaining compatibility with existing software.

Once we started looking at this, we found that it’s actually quite simple. As a user, the only thing you need to do is build a container image containing your VM image

FROM registry.gitlab.com/avassa-public/vm-orchestration:base

ADD noble-server-cloudimg-amd64.img
ENV VM_IMAGE=noble-server-cloudimg-amd64.img

As simple as that!

Once the container image is deployed to an edge site, the Edge Enforcer will configure KVM (The Linux hypervisor) to start the virtual machine.

Turns out that Windows at the edge doesn’t need to be any more complicated than that.

The Future of Application Management at the Edge

As we look ahead, the trend toward containerization at the edge will only accelerate. While VMs will remain necessary for some time, the long-term trajectory is clear: a container-first approach provides the agility, scalability, and manageability that modern applications require. By embracing this shift now, businesses can future-proof their edge deployments, reduce operational complexity, and unlock new opportunities for innovation. The future of edge computing is containerized—are you ready to make the transition?

Watch our Edge Field Day presentation to dig deeper into manageing VMs and containers in the Avassa Edge Platform.