← Back

#Postgresql

#Postgresql

Chirps tagged with #Postgresql

HawDB is an embedded graph database built in Rust that enables applications to efficiently store, query, and explore connected data using native Cypher and #PostgreSQL-compatible #SQL. It combines query optimization, reliable ACID transactions with crash recovery, optional full-text and vector search, and graph analytics in a lightweight, configurable package that runs directly within an application. Developers can experiment with the online playground

https://hawdb.ai/playground/

hawdb.ai

Playground — HawDB

A real terminal for HawDB: type Cypher or SQL, press Enter, see the result. Runs entirely in your browser via an experimental wasm32 build.

RobloxDB is Roblox’s in-house, next-generation ‌multi-tenant distributed database platform‌ that underpins every online data workload across the Roblox ecosystem, built entirely by Roblox’s dedicated database engineering team rather than relying solely on off-the-shelf solutions .

It is purpose-engineered to meet the platform’s extreme global scale requirements, supporting high-QPS, low-latency operations for tens of millions of concurrent users, while delivering strong consistency, multi-region high availability, and elastic scaling to keep services fast, reliable, and cost-efficient as Roblox’s user base and data footprint expand toward multi-exabyte volumes .

Core technical capabilities of RobloxDB include custom-built database engine enhancements (optimized indexing, storage formats, WAL, replication protocols, sharding, and query planner improvements), an advanced control plane that enables autonomous healing, zero-downtime schema updates, and seamless tenant migration across global regions, plus full integration with Kubernetes/Nomad container orchestration stacks paired with comprehensive observability frameworks . The team behind RobloxDB also actively contributes upstream improvements to open-source ecosystems including #PostgreSQL and RocksDB, extending the value of their custom engineering work to the broader database community .

The system is built on deep expertise in database internals (B-trees, MVCC, buffer management), consensus algorithms (Raft, Paxos), and low-level performance optimization, with rigorous automation for testing, CI/CD, rollout safety, and auto-scaling that allows the platform to operate reliably at massive scale with minimal manual intervention

https://careers.roblox.com/jobs/8129410

careers.roblox.com

careers.roblox.com

#PostgreSQL 19 Beta review uncovered a large number of unresolved, critical design and implementation flaws in the newly developed #SQL/PGQ property graph feature, ranging from broken dependency handling, insufficient locking mechanisms, incorrect permission checks to various crashes and unexpected runtime errors, that surfaced after a broad manual and AI-assisted code review.

Given the severity and volume of these unaddressed issues, combined with the upcoming release timeline for #PostgreSQL19, the #PostgreSQL Release Management Team reached a consensus with core contributors that the feature was nowhere near production-ready.

As a result, the community ultimately decided to fully revert the entire #PGQ implementation from both the #PostgreSQL 19 stable branch and the main development master branch, to eliminate time pressure and allow developers to rework the feature properly from scratch without being constrained by backwards compatibility requirements.

Learn more: https://www.postgresql.org/message-id/flat/dqa5mstx5mna3i7s23pdwl4m6bek7gqsgfccef44wjpswizufi%403aa6vzri3cat

Open Graph Image
www.postgresql.org

PGQ catalog representation and pg_dump support

← Newer Page 1 of 2 Older →