Focus

Cloud, Infrastructure, and Operations

Articles, podcasts, talks, and more about Cloud, Infrastructure, and Operations.

Managing Geopolitical Risks with Enterprise Architecture

IT leaders in Europe increasingly face questions about whether geopolitical developments – such as data protection disputes or trade conflicts between the EU and the US – threaten the viability of US cloud services. The risks range from price increases and legal uncertainties to potential usage restrictions. CIOs must not only ensure functional IT operations but also proactively assess external risks. This article demonstrates how enterprise architecture methods can help identify risks early and develop viable alternatives.

The Path to Heterogeneous Cloud Platforms

Digital transformation brings both challenges and opportunities for businesses. To create future-proof and flexible IT infrastructures, more and more companies are adopting multi-cloud or hybrid cloud strategies. This isn’t just about cost reduction, but also about aligning innovative services from different providers with regulatory requirements - particularly in the area of data protection. This article shows how targeted integration strategies can help you leverage the benefits of multi-cloud to optimize your existing infrastructure, reduce technical dependencies, and ensure long-term business success.

EU Data Act: The Beginning of the End for Cloud Monoculture?

Infrastructure in architectural documentation

Most people are familiar with the principles of good architecture documentation. However, the infrastructure often goes unnoticed or is only understood by a few “experts”. Especially with complex systems, it is important to record solution concepts and decisions at the infrastructure level of an application in order to be able to understand the architecture and further develop the software. This article is aimed at development teams that have difficulties integrating infrastructure topics into their documentation. It also offers insights for platform engineers and cross-functional teams who are responsible for their own infrastructure.

Socio-Technical Architecture as a Competitive Advantage

In today’s business world, technological progress is not just about increasing efficiency—it also unlocks entirely new business opportunities that would be unthinkable without these technologies. As a result, leveraging new technologies effectively has become a key competitive advantage. However, IT decision-makers are increasingly realizing that technology alone is not enough. Instead, companies must understand themselves as socio-technical systems, where technical systems and human interactions are deeply interconnected. This article explores why socio-technical architectures matter for IT decision-makers and how they can create a competitive edge for organizations.

Platforms, Teams, and APIs: How Do They Fit Together?

Common Approaches in the Field of Socio-Technical Architectures

In the age of digitalization and modernization, companies face the challenge of not only keeping up with technological advancements but also optimizing their organizational structures. Socio-technical architectures play a crucial role in this process by focusing on the flow of work, team collaboration, and the seamless integration of technical and social systems. This article explores two key approaches—Team Topologies and Domain-Driven Design (DDD)—and demonstrates how they can help organizations take the first steps toward consciously designing their socio-technical architecture.

Implementing GitOps without Kubernetes

“GitOps only works with Kubernetes as the operating platform!” This persistent misconception about GitOps might explain why it hasn’t become a standard workflow tool for software delivery yet. Most available GitOps tools require Kubernetes. However, GitOps fundamentally means using a version control system (VCS) like Git as an interface for a software agent to handle deployment and operational tasks within a target environment.

Compliance in hybrid operating environments

A socio-technical view

Is Platform Engineering the new DevOps?

When companies transition to the cloud, they anticipate an improved developer workflow, shorter release cycles, and more robust end-to-end team responsibility. However, even after the admin team delivers the new platform with all its capabilities, software teams may hesitate to adopt it and assume responsibility. But why? Is it not every developer’s desire to control the process from committing code to deploying it to production? One possible explanation is a lack of emphasis on the developer experience during platform conceptualization. This article aims to identify common issues and explore whether platform engineering can help to address them.

Build an internal development platform

We are currently breaking down our software into smaller pieces, such as SCS or microservices, and deploying them on cloud platforms to take advantage of their elasticity and flexibility. However, as our technology stack grows more complex, it has become increasingly challenging for our development teams to focus on business goals while also meeting operational requirements. In order to streamline the development process and provide easy self-service options, we are exploring the implementation of an internal development platform. To address these challenges, we have identified several open-source projects that can be integrated into our platform to enhance the development experience. Although this list is not exhaustive, it provides a solid foundation for creating a more effective platform. Our aim is to help development teams stay focused on business goals by minimising distractions and addressing operational needs.

Kustomize Enhancement with KRM Functions

Advanced Kustomize features

Kustomize’s strength lies in its simplicity, but now and then we have use-cases that need more flexibility and more dynamic behavior. For some of them, Kustomize offers additional features. But as with everything good, it should be used wisely. Otherwise, we fall into the trap of complexity and lose all the good parts of Kustomize.

Infrastructure Self-Service with Crossplane

Introducing container technology and Kubernetes to the development and deployment process allows developers to independently manage deployments of their services without needing to interact with the platform team. This is true for applications but not necessarily for external resources like databases which are probably still in the hands of the platform team because the access management and tooling differs. Crossplane could fill this gap towards a full self-service capability.

Managing Kubernetes Resources with Kustomize

An introduction to a template-free alternative to Helm

Happy without a Service Mesh

It does not always have to be a service mesh. Even microservices are not a good idea per se. For good decisions we have to get a deep understanding of the problem first.

Automating Dependency Updates with GitHub Actions

This describes setting up a GitHub Actions workflow from scratch to schedule automated dependency updates.

Scraping a Docker Swarm service with Prometheus

Kubernetes Probes

Kubernetes probes are for validating container health and readiness. But how should these probes be used? Do we need both? Can we share endpoints? This post tries to give some recommendations.

Cloud Computing and Carbon Footprint

Businesses around the world start to grow more conscious of environmental issues. Yet, although a decision made for an IT infrastructure is often one made for the long term, its carbon footprint is rarely taken into account.

Automating GitHub Pages Deployments with GitHub Actions

Integrating Logback with JLine

Logback is a popular logging library for Java. JLine can be used to create nice shell-like applications. Combining both is not as hard as it looks.

Prometheus Counters and how to deal with them

In this article, I want you to join me on my way to understand how Counters in Prometheus work, and how to query them to get the right information.

Structured Logging with Structured Arguments

A Playground for Testing OpenID Connect

This post describes how you can set up a development environment in order to play around with your OpenID client implementation. When running your application in a cluster, it can be difficult to test how it will behave behind a load balancer. One factor that can be particularly difficult to test is when you are communicating with an OAuth 2.0 or OpenID Connect server which expects that a request will be redirected back to the same application instance that it came from.

Secure Kubernetes Cluster

Containers and Kubernetes have greatly simplified the way we bring software into production. Applications can be packaged in container images and run on a Kubernetes cluster with 5000+ nodes. However, it should not be forgotten that, despite being encapsulated in containers, various aspects of safety still have an important role to play.

Carbon dioxide monitoring with Rust, InfluxDB and Grafana

Horizontal automation – a waste of time.

Automation of server administration tasks is a waste of time.

Per request debugging with Log4j 2 filters

Customers quite regularly call on me to support them when an application is not behaving as expected. On a development machine there are a lot of ways and utilities to analyse the situation. But troubleshooting the erroneous application behaviour in production environments quite often limits your tool belt to a minimum.

Hystrix – to make your fuse blow just in time

Case Study

DEVK: Standardizing Security and Compliance Across a Complex Software Landscape

New Primer: Cloud Migration

New Primer: Kubernetes Resource Management

Case Study

From idea to MVP in 8 weeks: An insurance platform for e-sports players

Case Study

A cloud platform for efficient commissioning processes in dentistry