Principles of Mechanical Sympathy Principles of mechanical sympathy for creating performant software: Memory access patterns, false sharing, the single-writer principle, and natural batching.

NEW POST

Modern hardware is fast, but software often fails to leverage it. @withcaer.com guides their work with mechanical sympathy. They distill this into principles: predictable memory access, awareness of cache lines, single-writer, natural batching

martinfowler.com/articles/mec...

The Last Quiet Thing Your possessions came alive. Now they won't stop talking.

This is an absolutely beautiful, glorious description of the products in our lives. The things that once were simple tools that met needs, now demand relationships, have opinions, and occupy cycles of our lives.

Do yourself a favor, scroll down and read this page
www.terrygodier.com/the-last-qui...

Tony Hoare (1934-2026) Turing Award winner and former Oxford professor  Tony Hoare passed away last Thursday at the age of 92. Hoare is famous for quicksort, ALGO...

RIP Tony Hoare, March 5 2026, age 92

blog.computationalcomplexity.org/2026/03/tony...

Should You Be A Carpenter? Is entering the "knowledge economy" today too risky given the field's increasing focus on AI?

I'm doing a new discussion series with Demetri Spanos on emerging AI topics. Ironically, the series art was made by hand by Anna Rettberg without generative AI :)

First episode is available now: www.computerenhance.com/p/should-you...

That’s awesome! Congrats.

CPUs are getting worse.



We’ve pushed the silicon so hard that silent data corruptions (SDCs) are no longer a theoretical problem.



Mercurial Cores are terrifying because they don’t hard-fail; they produce rare, but *incorrect* computations!

Have you seen the work Golang team has been doing in this area?

bsky.app/profile/swtc...

Why there’s no European Google? Why there’s no European Google? par Ploum - Lionel Dricot.

ploum.net/2026-01-22-w...

Updated JOL and heapdump-estimates in it specifically to better support JDK 25. We use this tool to quickly glance where are the opportunities in our real services. Streaming heap dump analysis allows doing this on multi-hundred GB heap dumps. Builds: builds.shipilev.net/jol/

Async-profiler has released an update that includes native lock profiling and a latency filter. Consequently, ap-loader, which wraps async-profiler in a platform-independent JAR, has also been updated: github.com/jvm-profilin...

“Floating-Point Printing and Parsing Can Be Simple And Fast”

The fastest known floating-point printer and parsing algorithms - fixed-width printing, shortest-width printing, and parsing, all in 400 lines of Go.

research.swtch.com/fp
research.swtch.com/fp-proof

A new way to call C from Java: how fast is it? lemire.me/blog/2026/01/17/a-new-wa...

How a 40-Line Fix Eliminated a 400x Performance Gap | QuestDB A deep dive into an OpenJDK commit that replaced slow /proc file parsing with a single syscall, revealing obscure Linux kernel internals and a 20-year-old optimization opportunity.

I did some OpenJDK commit exploration and I thought I could blog about it too. Here it is: questdb.com/blog/jvm-cur...

Quickly Inspect your Java Application with JStall - Mostly nerdless Learn about JStall, a tool that helps you quickly answer the question: "What is my Java application doing right now?"

I hope I'm not the only one who sometimes wonders: "What is my Java application doing right now?" When you don't see any output. This is where my tiny tool called JStall comes in, as you can read in my blog post at mostlynerdless.de/blog/2025/12...

My colleague Sanjay Ghemawat & I have done a fair bit of performance tuning of various pieces of code. We wrote an internal Performance Hints document ~2 years ago as a way of identifying some general principles & we've recently published a version of it externally.

Doc: abseil.io/fast/hints.h...

Mount Mayhem at Netflix: Scaling Containers on Modern CPUs Authors: Harshad Sane, Andrew Halaney

“Mount Mayhem at Netflix: Scaling Containers on Modern CPUs” from @netflixeng.bsky.social

netflixtechblog.medium.com/mount-mayhem...

How a Kernel Bug Froze My Machine: Debugging an Async-profiler Deadlock | QuestDB How I investigated and worked around a kernel bug that caused async-profiler to freeze my machine whenever I tried to use a profiler.

My machine froze when I connected async-profiler and I wrote an article about it.

It involves debugging a half-dead Linux kernel, performing live surgery on a frozen CPU with GDB, pulling off a questionable resurrection and other party tricks :)

Read the full autopsy: questdb.com/blog/async-p...

Here is my "Java Advent" article on nice and naughty cases of pattern matching, with nicer formatting than WordPress can muster, and the ability to leave comments without logging into WordPress: https://horstmann.com/un

Emily is really digging deep here.

Code formatting is messed up unfortunately.

Europe is under siege Menaced by Russia and China, abandoned by America.

Sadly I’m inclined to agree — Europe needs to get its act together, strongly support Ukraine and become able to defend itself, as America is not going to come and help this time www.noahpinion.blog/p/europe-is-...

High-Performance DBMSs with io_uring: When and How to use it We study how modern database systems can leverage the Linux io_uring interface for efficient, low-overhead I/O. io_uring is an asynchronous system call batching interface that unifies storage and network...

High-Performance DBMSs with io_uring: When and How to use it

Sharing is (S)Caring: How Tiered Compilation Affects Java Application Scalability Understand how Tiered Compilation impacts the scalability of Java applications in modern environments.

PR that addresses, Java JIT performance counters contention when multiple threads need to update them.

github.com/openjdk/jdk/...

redhatperf.github.io/post/method-...

#OpenJDK #Jit

Putin's plan to end the war in Ukraine.

Cartoon by Michael de Adder

This new AI world continues to frustrate the hell out of me. I'm seeing pull requests that clearly don't solve the problem, conversations that are automated nonsense, and fake engagement across all social media. No wonder people have no excitement about OSS work lately. ☹️

No Cords Attached: Coordination-Free Concurrent Lock-Free Queues The queue is conceptually one of the simplest data structures-a basic FIFO container. However, ensuring correctness in the presence of concurrency makes existing lock-free implementations…

No Cords Attached: Coordination-Free Concurrent Lock-Free Queues

Latency Profiling and Optimization - Dmitry Vyukov About Dmitry Vyukov Dmitry is a Principal Software Engineer at Google. He works on dynamic analysis tools for C/C++ (like Address/Memory/ThreadSanitizer), similar tools for the Linux kernel, and…

Latency Profiling and Optimization

Help--my LocalDate isn’t flattened! Just before my Valhalla talk at JFall, I found out that the early access build can flatten LocalDate arrays, but the latest version from Github cannot. Why? Serialization, of course. Read the gory details at https://horstmann.com/un