Performance news

Java Champion Gunnar Morling shares insights on building high-performance Java applications for data engineering. He discusses experiments with durable execution engines and the development of Apache Hardwood, a new, minimal-dependency Java parser for the Apache Parquet file format, offering lessons for developers and engineering leaders.

Shopify has launched GraphQL Cardinal, a new execution engine that uses a breadth-first approach instead of the traditional depth-first method. This change delivers significant performance gains, including up to 15x faster field execution, 6x lower garbage collection overhead, and improved latency for large-scale queries.

LinkedIn engineers faced recurring, short-lived outages where a key database would freeze and recover without leaving logs. Using advanced off-CPU profiling with eBPF, the team identified a subtle kernel lock contention issue as the root cause, resolving the mysterious system-wide freezes.

A new version of pg_sorted_heap, a PostgreSQL extension, has been released. It introduces physically sorted storage and integrated vector search. Version 0.14.0 adds official support for PostgreSQL 16 and is now available on the PostgreSQL Extension Network (PGXN) for easier installation and management.

AI brand discovery platform ChatFeatured migrated its analytics database from PlanetScale Postgres to a Postgres-compatible service managed by ClickHouse. The switch, completed in 30 minutes, reduced complex query times from 2.5 minutes to under one second, significantly improving performance for its AI-powered features.

The common practice of only measuring startup time for Spring Boot applications, especially with GraalVM or Spring AOT, is highlighted as a trap. While a quick metric, it fails to capture the full operational costs and complexities. Developers and architects need to consider a broader set of metrics for sound decision-making.