February 01, 2021

data migration and ETL

ETL stands for Extract, Transfer, Load. It is a process used to move information or data from one system into another environment. Data migratoin and ETL is similar as they both involve moving the data while data migration does not change the format, whereas ETL does.

The Transformation Process

This may include

  • standardization
  • deduplication (excluding and discarding redundant data)
  • verified
  • validated
  • sorted
  • aggregating of columns
  • translating coded values
  • encoded free-form values
  • joining / splitting column

Migration design

  • How data is extracted, held and verified
  • Mapping rules
  • How data is loading into new system
  • Recovery plans for each stage of migration
  • Schedule of actions

ETL Cycle

  1. Cycle Initiation
  2. Build Reference data
  3. lookup data (eg country codes, conversion rates)
  4. Extract
  5. Validate (cleansing)
  6. data type
  7. range and constraint
  8. code and cross reference check
  9. consistency check
  10. Transform
  11. Stage
  12. load into staging tables
  13. Audit reports
  14. Publish
  15. to production tables
  16. Archive

This process is pretty enlightening!

Read More: