Step 04 of 10 8-12 weeks· advanced
Step 4: Data Migration Design
Plan how data moves from mainframe storage (DB2 z/OS, VSAM, IMS, IDMS, flat files) to a modern relational target. Schema design and ETL strategy.
Recommended prompts
Use one of these to do the work in your IDE
Open the template to read it in full. Click Copy prompt to grab it (with your stack values pre-filled where they apply) — then paste into Claude Code, Cursor, or wherever you build.
Recommended skills
Drop these into Claude Code for this phase
Skills auto-trigger on the right kind of request. Install once; they apply to every prompt that fits.
Recommended MCP configs
Wire these tools into Claude Code first
MCP servers give Claude Code direct access to external systems (Jira, browsers, databases). Configure once.
When you're done
Verify these in your own work before moving on
This is a checklist for you to mentally tick off in your repo and IDE — the site doesn't track it, you do.
- Schema mapping complete for all major datasets
- ETL strategy chosen (big-bang / dual-write / CDC / trickle)
- Tooling chosen and validated on sample data
- Encoding strategy documented (EBCDIC → UTF-8)
- Sign / decimal conventions handled (packed / zoned decimal)
- Validation queries written
- Rollback strategy clear (preserve source until well after cutover)
Common pitfalls
What goes wrong at this step
- Trying to "improve" the schema during migration — migrate as-is; clean up later. Two changes at once are two bugs at once
- Ignoring EBCDIC encoding — surfaces as garbled characters days later
- Underestimating volume — "it's only a few terabytes" is a few terabytes you have to validate
- Skipping spot-checks — counts and sums match while specific records can still be wrong