Partitioning strategy for Oracle to PostgreSQL migrations on Azure
summary
Uses ora2pg to convert schemas, Azure Data Factory to run parallel copy jobs (five concurrent copy tasks), and PG Partman to auto-create and manage partitions. Demonstrates migrating a 170,000-row table into two partitions (101k / 69k) and a 300,000-row table into three 100k partitions, creating 60 partitions with Partman for a time-series table and pruning older partitions to three days. Data loads finish in parallel and indexes are created after the bulk load.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 6:15Oracle vs Postgres partitioning
- 8:01Steps to migrate partitioned tables
- 12:14Demo of migration end-to-end
- 22:30Important Resources (find in slides linked below)
related talks
25mMigrating VLDBs from Oracle to Azure Database for PostgreSQL
29mTales from the Field - Oracle to PostgreSQL migrations
24mMigrating from Oracle to Azure Database for PostgreSQL, Seamlessly
34mPartitioning and Data Migrations Using PostgreSQL as an Example
34mPartitioning and Data Migrations Using PostgreSQL as an Example
23mPostgres Partitioning Best Practices
01h 09mReplacing Oracle Features in PostgreSQL: HA, Functions & Extensibility
48mMigrating to PostgreSQL: Strategies, Sovereignty, and Success