This flow diagram shows you your options of our recommended solution patterns to migrate your on-premises data lake to Oracle Cloud. It helps you select a suitable pattern based on your organizational requirements.

You must first decide if you want to migrate your Big Data Cloud Service (BDCS) or Big Data Appliance (BDA) data lakes to Oracle Cloud using Hadoop clusters or without them. The following are the recommended patterns based on your choices.

Greenfield (Cloud Native Services)

Use this pattern for a new data lake implementation on Oracle Cloud. If you decide to use Hadoop clusters, you can leverage Oracle's Big Data Service or Hadoop workloads. If you choose not to use Hadoop clusters, you can build a data lake using cloud native data and AI services and leverage the ongoing Cloud Native Big Data pipelines.

Rehost (Lift & Shift Big Data Service)

Use this migration pattern and leverage it's lift and shift capabilities if you want to use Hadoop clusters and migrate your Hadoop workloads to OCI.

Rebuild (Cloud Native Services)

Use this migration pattern if your functionality directly maps to OCI cloud native services. You can also use the Rebuild pattern, if you're creating a new data lake on Oracle Cloud.

Replatform (Hybrid Model)

Use this hybrid approach with cloud native services if your requirements don't map directly to cloud native services. To use this pattern, you must first identify and map your functionality of Hadoop and cloud native services. Then identify which functionality goes where and adopt a hybrid approach to move your data based on your functionality mapping.