1. Spring and Java

>> Dissecting the CPU-Memory Relationship in Garbage Collection [norlinder.nu]

A deep exploration of garbage collection’s CPU overhead beyond the well-understood pause-time metric. The piece introduces a new Java API targeting OpenJDK 26 that allows teams to make more informed decisions about heap sizing and collector selection in production workloads.

>> How Project Leyden brought a new perspective [quarkus.io]

An insightful analysis of how Project Leyden’s class caching approach revealed previously hidden Java startup performance bottlenecks — from reflection-based compatibility layers to excessive ServiceLoader usage. The flamegraph-driven investigation is quite useful here.

>> Optimizations in Spring MVC [spring.io]

And a thorough benchmark analysis of Spring MVC performance across varying data set sizes, revealing that simple changes — read-only transaction annotations, HTTP header optimizations in Spring Framework 7.0.x, and virtual threads — can yield up to 20% throughput improvements for smaller data sets. Very cool.

Also worth reading:

Webinars and presentations:

Time to upgrade:

2. Technical & Musings

>> Knowledge Priming [martinfowler.com]

A practical proposal for treating project context as versioned infrastructure files that “prime” AI coding assistants before code generation, moving away from ad-hoc explanations toward a structured, repeatable approach. This is a compelling pattern for teams looking to get more consistent results from AI-assisted development.

Also worth reading:

3. Pick of the Week

And the new SOLID course is out:

>> Learn SOLID Principles

« Previous
Java Weekly, Issue 634
guest
0 Comments
Oldest
Newest
Inline Feedbacks
View all comments