OffNet Newsroom

Daily topic roundup

Database Technology

Wednesday, July 22, 2026 · 7 stories, curated & summarized — click any story for the source.

Share this roundup on LinkedIn Full edition
AWS Database Blog awsdatabase

AWS Blog: Multi-Region Active-Active API with Prisma and Aurora DSQL

The AWS Database Blog demonstrates an architecture for building active-active APIs across multiple regions using Prisma ORM and Amazon Aurora DSQL. This approach leverages the distributed capabilities of Aurora DSQL to handle data replication while Prisma manages the schema and query layer. The post outlines how to structure the application to maintain consistency and availability in a multi-region deployment.

  • Aurora DSQL enables active-active deployments without complex custom replication logic.
  • Prisma ORM abstracts the distributed data layer, simplifying multi-region query handling.
  • This stack reduces operational overhead for developers managing global API endpoints.
Read at source →
Planet PostgreSQL database

PostgreSQL and the 2038 Problem: Why Timestamps Matter Now

An analysis of how the Year 2038 problem threatens database integrity by exposing risks in timestamp handling across critical operations. The piece highlights that modern cloud-native systems rely heavily on accurate time tracking for sessions, replication, and audit trails. It argues that databases preserve time as an operational reality, making this a fundamental infrastructure concern rather than just a theoretical edge case.

  • Timestamps underpin login sessions, token expiry, and distributed event ordering in cloud apps.
  • Replication timelines and backup retention schedules are vulnerable to time calculation errors.
  • Audit histories and certificate validation depend on continuous, accurate timekeeping.
  • Database systems must handle time as a core operational dimension to prevent data corruption.
WORTH QUOTINGTime Is Core Infrastructure“Databases must handle time as a core operational dimensionto prevent data corruption.— Tech News Analysis
Read at source →
Planet PostgreSQL database

Paul Jungwirth reviews TQuel paper ahead of Postgres 19 temporal features

With UPDATE/DELETE FOR PORTION OF slated for Postgres 19, Paul Jungwirth examines the 1993 TQuel paper to understand temporal relational algebra. The review focuses on algebraic identities for temporal operators, which are critical for query planner optimizations. By leveraging these identities, the planner can transform queries into more efficient execution shapes, such as pushing filters before joins.

  • Postgres 19 will support UPDATE/DELETE FOR PORTION OF, enabling native temporal data manipulation.
  • TQuel, a 1993 Ingres extension, provides foundational algebraic identities for temporal operators.
  • Understanding these identities helps explain how query planners optimize temporal query execution.
  • Filtering rows before joining tables is a key optimization strategy derived from these algebraic rules.
  • This historical context informs the design and performance expectations of upcoming Postgres temporal features.
TRADE-OFFTQuel vs Postgres 191993 TQuelFoundational temporal algebraHistorical research paperIngres extensionPostgres 19Native implementationUPDATE/DELETE FOR PORTIONProduction readyvs
Read at source →
AWS Database Blog awsdatabase

AWS RDS AI forensic analysis automates database incident response

AWS introduces a serverless method for continuous forensic artifact collection on Amazon RDS and Aurora, capturing point-in-time snapshots of database internals. These snapshots are stored in Amazon S3 to create a time-series record that AI tools can analyze instantly. This approach replaces hours of manual investigation with rapid, automated analysis.

  • Automated forensic collection reduces incident investigation time from hours to seconds.
  • Point-in-time snapshots capture database internals for historical AI analysis.
  • Serverless architecture eliminates operational overhead for continuous monitoring.
  • Stored artifacts in S3 enable rapid retrieval and conversational AI debugging.
Read at source →
AWS Database Blog awsdatabase

Nubank migrates critical payments to Aurora PostgreSQL with 1,900x query gains

Nubank moved its mission-critical payment infrastructure from self-managed PostgreSQL to Amazon Aurora PostgreSQL-Compatible Edition to resolve scaling and operational challenges. The bank established strict evaluation criteria for the new database solution, leading to a successful migration that significantly improved performance. Specific queries saw performance improvements of up to 1,900x, demonstrating the platform's capability to handle high-volume financial workloads.

  • Aurora PostgreSQL delivered up to 1,900x query performance improvements for Nubank's specific payment workloads.
  • Migration strategy included rigorous evaluation criteria to ensure operational efficiency and performance gains.
  • Self-managed PostgreSQL limitations drove the need for a managed, scalable database solution.
  • The move supports high-scale payment infrastructure with reduced operational overhead.
BY THE NUMBERSNubank's Query Performance Gain1,900xQuery performance improvement ratioAurora PostgreSQL vs self-managed for payments
Read at source →
Planet PostgreSQL database

pg_hardstorage enables PostgreSQL backup migration with preserved recovery strategies

pg_hardstorage facilitates migrating PostgreSQL backup solutions by maintaining existing operational workflows, retention policies, and compliance requirements. The tool supports a gradual transition where legacy repositories remain active while new backups are simultaneously written to the new pg_hardstorage system. This approach ensures operational continuity and preserves confidence in recovery procedures during the shift.

  • Migrate backup systems without disrupting retention policies or compliance standards
  • Run pg_hardstorage alongside existing repositories for a safe, gradual transition
  • Preserve operational runbooks and recovery procedures during the migration
  • Avoid single-point failures by keeping historical backups accessible during transition
Read at source →
AWS Database Blog awsdatabase ↺ since 07-21

SQL Server 2025 on RDS calls Bedrock directly from T-SQL

SQL Server 2025 on Amazon RDS now supports invoking Amazon Bedrock foundation models directly within T-SQL scripts via sp_invoke_external_rest_endpoint. This architecture eliminates the need for external middleware layers, allowing AI logic to run inside the database engine. The approach aims to reduce latency and streamline workflows by keeping data and AI processing in the same environment.

  • Use sp_invoke_external_rest_endpoint to call Bedrock models directly from T-SQL without middleware.
  • Reduced latency results from eliminating external application layers for AI inference calls.
  • Simplifies architecture by embedding agentic AI patterns directly into database workflows.
  • Enables AI capabilities to access database context natively within stored procedures.
Read at source →
↑