Box Plots, Or Better! 5 Reasons Why Box Plots are the Better Default Choice for Visualizing Performance stefan-marr.de/2024/06/5-reas… #performance #benchmarking #visualization #paperWriting

👋🏻 I’m at GopherCon Europe today and tomorrow. Come and say hello if you want to talk about go profiling, execution tracing, runtime metrics, frame pointers, performance, otel or anything else :). #gopherconEU #golang

I've just booked my trip and I'm so excited! If you want to talk about PHP performance, get your ticket and meet me in Lille 👋

git worktree (this is bonus comic #3 for "How Git Works" wizardzines.com/zines/git, covering some topics that I didn't talk about in the zine) permalink: wizardzines.com/comics/git-wor…

+1 I'll be at #DASH2024 in NYC as well! Happy to talk Ruby profiling, performance and whatnot!

Here are the slides for my GopherCon Europe talk: How to Win Frames and Influence Pointers docs.google.com/presentation/d… #gopherconEU #golang

When I first started writing this specific documentation socketry.github.io/async-websocke… nothing worked: Puma and Rails didn't support streaming responses. However, today I can write this documentation and it works out of the box with Rails 7.1+ and Puma 6+. And of course, Falcon too.

Really cool talk on using the Datadog, Inc. Continuous Profiler to investigate and fix #ruby performance issues in production :) Much better than my usual synthetic-ish examples of the tool haha

Spotlight on Felix Geisendörfer talking about #golang execution tracing: How to Win Frames and Influence Pointers youtu.be/0f8JnJhC0-0?si… #gopherconEU

My colleague Nick just dropped his first blog post: "The one-instruction window." It's a thrilling dive into one of the debugging adventures behind our go1.21 contributions to Go's execution tracer. Spoiler: Nick uncovered a very subtle compiler bug! 😱

youtu.be/Vu372dnk2Ak Great CCC with Ben Titzer talking V8 vs C2 Sea-of-Nodes. Reg Alloc, Scheduling, Safepoints, reducible, deopt, all over the map. Lots of WASM. Tiered compile. Multi-level IRs. Cost models, optimization. microbenchmarks

Hello, do you run a Rails application and for some reason don't use jemalloc? Clear your calendar for the next hour, and install jemalloc in production. Here's my latest retainer client saving almost 50% of their memory usage by making a 3 line change to use jemalloc.

Nate Berkopec on the topic, if you are using mini_racer and jemalloc, be sure to update to the latest version of the gem.... this one was a very surprising one: github.com/rubyjs/mini_ra…

After reading a beautiful conversation about number of threads in Rails(github.com/rails/rails/is…), I've dug a bit deeper and got so much cool info on Ruby's GVL, hopefully could convert it to the article later

🚀Do yourself a favor and use jemalloc. Yellow dotted line represents the Ruby process with -7 days offset (before introduction of Ruby 3.3.4 image with jemalloc from Fullstaq) #Ruby #RubyOnRails

Monthly shout-out to Shopify, Ted Thetnaungsoe, Gautam Rege, Mark Triggs and my other sponsors this month! Your contributions help ensure I can keep working on JRuby and improving the JVM language ecosystem. You can sponsor my work at github.com/sponsors/headi…

One of my favorite features of the Continuous Profiler we're building at Datadog, Inc. is Custom Context. It allows you to break down your resource consumption, like CPU usage, by any dimension. You can add anything you want: Customer ID, Rule ID, Country, etc ...