Examining an overview of the AWS multi-account environment

There are many variations of multi-account strategies that are valid. Multi-account implementations can vary based on the organizational and technical needs of a customer. For the purposes of this chapter, we will focus on a basic multi-account strategy, focusing on only the accounts that are most relevant to a machine learning workload using Amazon SageMaker. We don't explicitly call out accounts (such as security or logging) because they are already well defined in the context of AWS governance practices. Figure 14.1 illustrates the general, high-level accounts we will use to discuss the concepts in this chapter.

Figure 14.1 – Example of AWS accounts and SageMaker features

Using Figure 14.1 as an example, the following AWS accounts may be used as part of an end-to-end ML Lifecycle. Please keep in mind that account naming and resource placement may vary considerably across implementations...