Getting started checklist
During your data journey with cloud-scale analytics, you'll find there are multiple stages in your adoption lifecycle. This section provides a quick getting started checklist to help you adopt your scenario in stages. These stages are:
- Stage 1: First landing zone
- Stage 2: Extra data domains
- Stage 3: Improve consumption readiness
- Stage 4: Critical governance components
Stage 1 - First landing zone
- Define your first use case(s)
- Deploy your first data management landing zone
- Deploy your first data landing zone
- Define your first (ingestion) pattern (for example, batch parquet)
- Develop your first data product (ingested raw, abstracted to product)
- Determine 'just-enough' governance
- Define metadata requirements (application information, schema metadata)
- Register your first data consumer (manual process)
Stage 2 - Extra data domains
- Refine your target architecture
- Deploy more data landing zones
- Extend with second, third and fourth data products
- Realize your data product metadata repository (database or excel)
- Implement your first set of controls (data quality, schema validation)
- Realize your consuming pipeline (taking input as output)
- Establish data ownership
Stage 3 – Improve consumption readiness
- Implement self-service registration and metadata ingestion
- Offer other transformation patterns (transformation framework, ETL tools, etc.)
- Enrich controls on the provider side (glossary, lineage, linkage)
- Implement your consuming process: approvals, use case metadata, deploy secure views by hand
- Establish your data governance control board
Stage 4 – Critical governance components
- Apply automation (automatic secure view provisioning)
- Deploy strong data governance, set up your dispute body
- Finalize your data product guidelines
- Define your extra interoperability standard
- Develop your self-service data consumption process
- Develop your data query, self-service, catalog, lineage capabilities, etc.
- Develop more data marketplace capabilities.
Summary
These four development stages allow you to set up a minimal viable product in stage one, learn, and iterate into stage two. Throughout your staged approach, you grow in maturity for creating a self-service, scalable, and governed platform.