Priya Raghunathan
Data Platform Lead
Priya has migrated three petabyte-scale warehouses without a maintenance window and has the scar tissue to explain why you probably should take one.
- DRM-free PDF and EPUB, yours to keep
- Instant download the moment you check out
- Free updates for the life of the edition
The Warehouse Migration
Moving Petabyte Analytics to BigQuery and Redshift
Multi-cloudAdvancedData
Warehouse migrations fail for boringly consistent reasons: undocumented downstream consumers, semantics that shift under the new engine, and a cutover plan that assumes correctness can be verified after the fact.
Priya Raghunathan gives you the sequence that works — inventory consumers first, pin semantics with differential testing, migrate in shadow, and only then move the readers. She covers both BigQuery and Redshift, including where their cost models will change your schema.
Includes a full differential-testing harness design you can adapt to your own dialect pair.
What you'll learn
- Inventory every downstream consumer before you move a table
- Catch semantic drift with differential testing across engines
- Design a shadow-migration cutover with a real rollback path
- Adapt schema and partitioning to each engine's cost model
Table of contents
- 01Why These Migrations Fail26 pp
- 02Consumer Inventory34 pp
- 03Semantic Drift Between Engines44 pp
- 04Differential Testing at Scale46 pp
- 05Partitioning and Cost Models40 pp
- 06Shadow Migration42 pp
- 07Cutover and Rollback36 pp
7 chapters · 398 pages total
Details
- ISBN
- 978-1-959321-05-2
- Edition
- 1st
- Pages
- 398
- Published
- Formats
- PDF, EPUB
- Platform
- Multi-cloud