Architecture
While there are many architecture possibilities, it will always be divided into four buckets: ingest, store, process, and server. The components inside these buckets can change based on individual preferences. Components are often chosen based on the team’s skills – for example, a team that has experts in Databricks might prefer to use it for ingestion. Figure 14.1 shows one possible architecture for this workload:
Figure 14.1 – Advanced analytics with Databricks and Azure Machine Learning
Take a moment to study this architecture diagram; then, move on to the next section to learn more about the components involved and their functionality.
