Organizations are drawn to migrate their systems into the cloud, but in practice, cloud transformations do not always succeed in achieving the advantages often promised by cloud providers, such as scalability, availability, and cost effectiveness.
We give an overview over Kevin Hoffmann's fifteen factors for cloud-native applications and how they help us achieve success in our cloud migrations.

Target Audience: Architects, Developers, Operations, DevOps
Prerequisites: Basic understanding of web services
Level: Basic

Extended Abstract:
You think you are already reaping the cloud's harvest by lifting and shifting your apps to some cloud-provided VM or container infrastructure?
You've already read about Heroku's (original) twelve factors, maybe even skimmed though Kevin Hoffmann's book, and think your app is good enough, because you are using Git, have a CI pipeline, and someone in the team once ran the app in a Docker container?
Or you think your team isn't going to the cloud any time soon and this advice does not apply to you anyway?
Maybe you're right. Chances are, you are in for a surprise, though.
The original twelve factors for cloud-native application development are (reasonably) well established and widely agreed upon. Yet, in practice, there appear to be some misunderstandings regarding their meaning, and we see cloud migrations fail precisely due to reasons that could be avoided by following these factors correctly. The extension to fifteen factors by Kevin Hoffmann is less widely known, but no less important for modern, cloud-native application landscapes, especially since this extension also revisits and updates the original twelve factors.
The goal of this talk is to disseminate and discuss the fifteen factors, as well as to emphasize their interrelations and embed them in the greater context of modern application development, in order to help reduce pain for many teams tasked with bringing their software to the cloud.

At Tchibo we wanted to reduce our server and energy consumption with our product development teams. But why would we care? We all like fast snappy development and test systems. And our shop needs to survive Black Friday’s shopping traffic. Annual Google Cloud consumption forecasts do not trigger us to consume less. But when we started to show teams their related carbon dioxide footprint, we created a feedback loop that will help us to become better CO₂-developers.
As developers, we have an impact - to program CO₂ reduction.

Target Audience: Everyone, Developers, DevOps People
Prerequisites: None
Level: Advanced

Extended Abstract:
We will present the idea and first measurements on the server side. The planned client side measurements are still work in progress and very difficult to perform. Nevertheless is the client side induced CO₂ consumption at large websites probably a large lever for reduction.

Mehr Inhalte dieses Speakers? Schaut doch mal bei sigs.de vorbei: https://www.sigs.de/autor/johannes.mainusch

High traffic during business hours, no traffic at night, weekends and vacations, multiple teams, and several staging environments - these characteristics of a typical enterprise application. Pay-as-you-go, "scale-to-zero" and managed services make serverless architectures appealing for enterprise applications.
On-premise, on the other hand, you get the maximum flexibility and full access to machines with less automation and so more plumbing.
I will compare both approaches with focus on architecture and answer your questions in real time.

Target Audience: Developers, Architects
Prerequisites: Basic Cloud and Java Knowledge
Level: Advanced

Great engineers often use back-of-the-envelope calculations to estimate resources and costs. This practice is equally beneficial in Machine Learning Engineering, aiding in confirming the feasibility and value of an ML project. In my talk, I'll introduce a collaborative design toolkit for ML projects. It includes Machine Learning Canvas and MLOps Stack Canvas to identify ML use cases and perform initial prototyping, thus ensuring a business problem can be effectively solved within reasonable cost and resource parameters.

Target Audience: Architects, Developers, Project Leader, Data Scientist
Prerequisites: Basic knowledge in machine learning
Level: Advanced

DevOps isn't dead yet and it will not die in the future. Platform engineering comes to leverage DevOps practices horizontally to support developers, operations and those in between. In this session, I will give you a perspective on how to move on from DevOps to Platform Engineering, how to design and shape your internal platform and build a vibrant community sharing best practices and enabling each other to overcome faster issues. In the end you will understand how we can reduce the cognitive load for dev teams to focus on features.

Target Audience: DevOps, Developer, Manager
Prerequisites: Experience in DevOps and an understanding of the pain of silos
Level: Advanced

Extended Abstract:
Platform Engineering is often seen as the evolution of DevOps. However, it is yet a new discipline using DevOps practices but focusing on the entire chain and breaking silos. I will show details from my experience on my job as well as 3 years of release engineering at Kubernetes where it wouldn't be possible to release without a proper platform.