This article is a follow-up to a series where I provide a concise, pragmatic overview of the steps involved in deploying the required infrastructure for installing your first SAP system in the cloud. This post gives an overview of all the elements and guiding principles that we should consider before starting to provision resources in Azure.
In the previous article (Day 0. Getting started) we provisioned an Azure account. With this in place, if you log in, you should be able to spin up a new virtual machine in about 5 minutes, or less if you script it.
Planning. What is a Landing Zone? Define goals. Define target architecture
But hey, we want to give some structure to what we deploy so that we can better manage resources afterwards, apply fine-grained access controls, have an overview of costs by resource, etc. This is what following best practices means, right?
This is a list of resources that we recommend you read in detail, to get a good idea of what aspects should be considered when planning a deployment in Azure:
- What best practices are the best? let us get started by looking at the Azure Well-Architected Framework: it is a set of guiding tenets that you can use to improve the quality of a workload
(Microsoft Azure Well-Architected Framework – Azure Well-Architected Framework | Microsoft Learn) - You should also get familiar with the Landing Zone concept: An Azure landing zone is an environment that follows key design principles across eight design areas.
- It is also important to consider whether you want to use Infrastructure as Code (IaC)– this would need another article (or more) on its own, let us just say that IaC has tons of benefits, that it requires some discipline and occasionally some “ramp up” time.
- Project management best practice tells us that we should identify business objectives, align them with technology initiatives, break them down into smaller projects and define KPIs so we can measure progress.
- As a final step, we should design our architecture (resource organisation, permissions, network, etc.) considering all the above. And, of course, we should think about what happens the day after we have successfully migrated our workload to the cloud – you will need someone to “operate” it, to deal with any requests or incidents that may arise, but also to continually optimise costs and performance, regularly check security, etc.
Lots to think about here :), our suggestion is not to over-engineer things: make a list of topics that are important to you, get it prioritized and document the outcome, so you can use it as a reference in the future.
In the next post of this series, Day 2. theory is good but, can we please get started? The ‘plumbing’, we will create the required infrastructure such as networking, identity and access management, monitoring and additional components that are part of a typical landing zone deployment.
