Public Sector

We've had the pleasure of working with UK and overseas central and local government departments, including Healthcare (NHS and Foundation Trusts), Defence, Education (Universities and colleges), many of the main Civil Service departments, Emergency Services; also public-owned corporations including the BBC, Bank of England, Ordnance Survey, and regulatory bodies such as Ofgem.

We are registered on Crown Commercial Service’s (CCS) Dynamic Purchasing System (RM6219 Training and Learning) and also with numerous tender portals such as Ariba, Coupa and Delta E-Sourcing.

Read more...

Graduate Training Schemes

Framework Training has a strong track record of providing a solid introduction into the working world for technical graduates across myriad industries. We provide the opportunity to learn and gain valuable hands-on experience in a supportive, friendly and sociable training environment.

Attract & retain the brightest new starters

We know it is vital for our clients to invest in the future of their talented grads; not only to provide them with high-quality, professional training essential for their roles, but to embed them within the organisation’s culture and guide them on the right path to a successful career.

After all, your new hires could well be the next leaders and their creative ideas and unique insights are invaluable to your business.

Read more ...

Learning & Development

Our unique portfolio of high-quality technical courses and training programmes are industry-respected. They’re carefully designed so that delegates can seamlessly apply what they’ve learnt back in the workplace. Our team of domain experts, trainers, and support teams know our field — and all things tech — inside out, and we work hard to keep ourselves up to speed with the latest innovations. 

We’re proud to develop and deliver innovative learning solutions that actually work and make a tangible difference to your people and your business, driving through positive lasting change. Our training courses and programmes are human-centred. Everything we do is underpinned by our commitment to continuous improvement and learning and generally making things much better.

Read more...

Corporate & Volume Pricing

Whether you are looking to book multiple places on public scheduled courses (attended remotely or in our training centres in London) or planning private courses for a team within your organisation, we will be happy to discuss preferential pricing which maximise your staff education budget.

Enquire today about:

  • Training programme pricing models  

  • Multi-course voucher schemes

Read more...

Custom Learning Paths

We understand that your team training needs don't always fit into a "one size fits all" mould, and we're very happy to explore ways in which we can tailor a bespoke learning path to fit your learning needs.

Find out about how we can customise everything from short overviews, intensive workshops, and wider training programmes that give you coverage of the most relevant topics based on what your staff need to excel in their roles.

Read more...

2025 State of Containerisation

Explore 2025’s containerisation landscape — from serverless models to AI-driven operations. Discover how Docker, Kubernetes & Edge Computing are evolving.

October 20th, 2025

Containers and containerisation in late 2025 have now moved beyond basic adoption to widespread usage in many / most operational environments. Indeed, it is no longer the preserve of the experimental or fast moving start up but is now the bread and butter of the well-established enterprise. This level of infiltration has been fuelled by multiple drivers including the adoption of serverless containers, the demand for multi-cloud architectures and the ever-growing impact of edge computing. In this blog we will look at the key containerisation trends in 2025, consider briefly the technology landscape and conclude by looking at the impact of AI and Machine Learning on the containerization world.

Key Trends in 2025

Serverless Container Model

During 2025 many organisations have moved towards a Serverless Container Model. In a Serverless model customers can utilize different cloud capability types without having to provision, deploy and manage either hardware or software resources. They of course still need to provide the application code and / or data and this is where the container aspect comes in. Containers can be configured as required and then deployment onto a serverless platform. This means that customers do not need to manage the actual servers or compute infrastructure that the containers are running on. They can merely create the container and deploy it into the serverless offering.

For example, AWS now offers a Serverless Container product, Fargate, which enables companies to run containers without having to manage EC2 servers or clusters. In turn Google also offers a serverless containers option with Cloud Run and Microsoft Azure offers Container Instances so their users can run containers in the cloud without managing any servers.

In many cases charges are based on the containers being deployed rather then on compute usage – another attraction of this model.

Dominance of Docker

Docker (launched back in 2013) is still the leading containerisation platform in use today. It is well known for its simplicity and stability. In many ways Docker is the face of Containerisation and for many when they think of a container solution they think of Docker. One analysis suggests that in 2025 Docker has been deployed 47% of companies with at least 1,000 hosts have Docker in full production meaning that larger companies are taking Docker adoption extremely seriously. That is not to say that Docker has it all its own way, alternatives such as Podman and containers are starting to gain traction particularly in security sensitive environments.

Dominance of Kubernetes

Whilst Docker may be the face of containerisation in 2025, Kubernetes is the Conductor ensuring that the orchestra of containers plays the tune that the customer wants to hear. Indeed in 2025 the Cloud Native Computing Foundation (CNCF) stated that 96% of organizations either use Kubernetes or are actively evaluating it Kubernetes is thus the undisputed leader in container orchestration. It has evolved into a stable, well established, and very important platform for running enterprise-wide container-based systems.

Alternatives to Kubernetes

Above we have said that Kubernetes in the dominant orchestration systems in the container world – and it is – at the moment! However, another trend in 2025 has been the rise of alternatives to Kubernetes. This is because of Kubernetes complexity, which had led many organisations to seek out simpler, easier to use and lighter weight solutions such as Fly.io, Hashicorps’ Nomad and AWS’s App Runner. Interestingly Docker Swarm also seems to be experiencing renewed interest – although not exactly a revival. This is in part due to its simplicity and ease of use and of course its native integration with Docker.

Cloud Native Containerisation is the New Normal

In 2025 the majority of Kubernetes clusters are now hosted in the cloud. This is illustrated by the ‘Kubernetes in the wild’ report by the Dynatrace organisation which found that 2 out of every 3 Kubernetes clusters are hosted in the cloud. This indicates that cloud-native computing and containerised applications are the standard for modern cloud-based systems. That is cloud native containerisation is the new normal!

Multi Cloud

This trend follows on from the previous one, if the majority of Kubernetes clusters are in the cloud, then what about those clouds? There are different cloud offerings being presented to many / most organisations whether that is due to differences in internal systems, versus public infrastructure to edge computing environments. Therefore, the need to be able to deploy applications across multiple cloud systems becomes paramount. The use of containers and orchestration systems such as Docker and Kubernetes can insulate an organisation from the differences between one cloud system and another.

Edge Computing

In a previous blog we said that “Edge Computing is the idea that data processing functions should be placed closer to the source of that data rather than in some remote processing facilities. The source of the data may well be devices such as smart vehicles, domestic home products or other devices which might be labelled as part of the Internet of things (Iot).” As the world moves ever closer to a fully interconnected reality, Edge Computing will only become more and more important.

Although the trend for edge computing is to use lighter weight orchestration solutions such as K3S and MicroK8s, containerisation is still key to the growth in this area.

Security

The container security market has been growing rapidly in 2025, driven of course by both the widespread adoption of containers and the complex security challenges that they present. This is demonstrated in several ways. For example, with compromised open-source images posing a growing risk, there is now an increased emphasis on ensuring that these buildings blocks are safe and secure with only using trusted images being used.

Of course, it is not just the basic images which are a cause of the problems. Bloated or non-minimal (but easy to access) container images create a larger opportunity for a security breach, and issue made worse but the rapid proliferation of container images.

Another issue is that security misconfigurations within containerized solutions are widespread. For example, one report stated that in 2022, 44% of organisations were running 71% or more of their workloads allowing root access!

However, identifying trusted sources may not be trivial and numerous tools are becoming available to scan images and generate reports on their trustworthiness. For example, Open-source container-security tools like Trivy and Clair automate the scanning process and generate detailed reports with severity ratings. Indeed, security is ‘shifting left’ so that vulnerability scanning and threat modelling can be integrated early into the development process.

Finally, threat detection is becoming more sophisticated with the use of AI and ML tools to analyse and identify anomalies. An example of tools working in this area is Vectra with its Ai based threat detection system.

Technology Landscape

A Maturing Market

Docker was released in 2013 and Kubernetes in 2014, in the just over 10 years since, these systems (and others like them) have matured into well established and solid suites of tools. Their usage in the IT industry as a whole has reached almost ubiquity and they are now common across all software industries. Their adoption in other industries such as manufacturing and XX is somewhat lower but still appears to be growing rapidly so that it is likely these other industries will catch up with the IT industry soon. The global software containers market is estimated at $4.5 billion in 2025 by future Market Insights and is projected to reach over $13 billion by 2035, indicating significant growth over the next 10 years.

Docker and Kubernetes Dominance

As stated above, the container world is dominated by Docker and Kubernetes in 2025. However, managing and operating Kubernetes remains a complex challenge for many organisations. This is probably a barrier to non-IT markets and certainly a headache even for those within the IT industry. How this will affect Kubernetes in the long run remains to be seen, but it is certainly a challenge.

CTA Banner encouraging people to learn about our Containerisation with Docker and Kubernetes trainingLogging and Monitoring

Logging and monitoring are critical within a dynamic containerised system, yet still in 2025 consistent visibility of the systems is still difficult and requires constant attention.

Containers and AI/ML

Within the computing world it seems that AI is the elephant in every room! The world of Containerised systems is no different, although perhaps there is more subtlety to the role of AI with respect to containerisation than might at first be thought. There are at least three ways in which AI and ML (AI/ML) might impact on Container systems, for example with regard to Container Management, Container Security and as an application area for Containerisation.

AI/ML for Container Management

AI/ML can be used to enhance the performance and reliability of the container infrastructure itself. For example, AI/ML can be used to predict workload resource requirements which can then be used to (potentially automatically) optimize container scheduling and placement. Thus, high priority workloads can get the resources that they need while reducing cloud costs by improving overall cluster efficiency. Although it is still early days with such technology most cloud infrastructure providers are investing heavily in such tools.

In addition, a new term AIOps (or AI for IT Operations), is being used to transform the way in which operations activities are performed. AIOps aims to automate and improve IT operations using machine learning and other AI techniques by analysing IT data in real-time.

It is being used to provide intelligent automation and predictive analytics within Kubernetes operations. These AI/ML models can help to identify occurring and potential issues before they occur and automate incident responses thereby reducing the average time to resolve them.

AI/ML is even being integrated into developer-oriented tools such as Docker Desktop to provide intelligence guidance to help developers building applications.

AI/ML for Container Security

Instead of relying on static rules, AI/ML is also being used to help with security within a containerised system. It can be used to perform proactive threat detection, automated vulnerability management and incident responses.

To do this AI/ML systems can analyse container behaviour to create profile of ‘normal’ activity., Any variation from this activity such as unusual network traffic or suspicious system calls can be flagged as a potential threat. The affected container can then be quarantined, or malicious traffic can be blocked etc.

For example, Falco is an AI powered threat detection engine that can be used to monitor system calls and detect unusual network connections etc. Another example is a ML-based log analysis system analysing login attempts looking for suspicion login attempts.

Containers for AI / ML workloads

Not only can AI/ML help the container world, but in return Containerised solutions can be beneficial to the AI/ML world. Containers can be used to help develop reproducible environments, avoid the old ‘it works on my machine’ problem. They can produce, for example, a Docker image, with Python, and the correct version of key libraries such as pandas, numpy, PyTorch, SciPy into a single image that can be shared.

Using containers and orchestration tools such as Kubernetes, containerised AI/ML solutions can be scaled out across many nodes allowing for large scale model training and high-volume processing of data.

In this area a rich ecosystem of containerised tools has emerged including Jupyter Docker Stacks, MLFlow containers and LLM (Large Language Model) containers such as Ollama.

Summary

The world of container-based systems has matured over the last 10 years to become one of stability and reliability. However, this world is still only starting to feel the effects of AI and Machine Learning on what can be done to help manage the complexity of these systems and ensure their security. The application of AI within this sphere is only likely to increase so what this space. What is clear is that at least for the next few years the impact of containerised systems is only likely to grow and grow.

Share this post on:

We would love to hear from you

Get in touch

or call us on +44 (0) 20 3137 3920