Episode

Accelerating PL/pgSQL Code Conversation When Migrating to Postgres

with Deepak Mahto

POSETTE: An Event for Postgres 2024

Video of a conference talk about accelerating PL/pgSQL code conversation presented by Deepak Mahto at POSETTE: An Event for Postgres 2024. Most of the adopters of PostgreSQL are folks moving from another database and code conversion or pl\pgsql development is a critical path. It is a session, we will walk through curated learning I had moving customers and large code bases to PostgreSQL with pl\pgsql.

Topics to cover:

  • Subtle art of code conversion, conversion is art not matching code always.
  • Key difference in PostgreSQL that is unknown to all assessment tools but critical.
  • How to migrate not only for primary instance but for Read replicas as well
  • Key Differentiated feature of plpgsql and how to leverage it.
  • Leverage newer features added in PostgreSQL like alias, unique on null and more.
  • Avoid surprises around partitions like missing primary keys in partition columns and more.
  • Why not to Ignore nulls and empty string comparison and what impact it can bring.
  • How to Handle Packages in Oracle and how to accelerate its migrations.
  • Build Code planning and uncover dependency at start of projects.
  • Data type mapping and how to automate it using ora2pg. And more.
  • Key Takeaways for the participant

Understand pl\pgsql best practise from a migration perspective. Uncover hidden caveats with code conversion. Tips to accelerate code conversion.

Deepak likes to call himself a Database Guy who works in database development, performance tuning, consulting and migration. He was born and brought up with Oracle and PostgreSQL databases. Deepak has spent the last 7 years actively working on database migration to Cloud. Passionate about PostgreSQL and all Cloud managed database services. Actively working on seeking new migration patterns for solving complex migration challenges when migrating to PostgreSQL.

Chapters

  • 00:00 - Beginning of Deepak's talk
  • 01:01 - Distribution of Database Migration Efforts
  • 02:34 - Why PL/pgSQL
  • 07:12 - Code conversion needs planning
  • 11:32 - Empty string vs. Null
  • 13:51 - Mapping data types
  • 17:31 - Extensions that help with code conversion

Connect

Azure
Azure Database for PostgreSQL