Performance analysis tool which makes it easy to use Intel Processor Trace to visualize every function call made leading up to an event of interest with nanosecond precision. I designed and implemented this (along with an intern I mentored) for Jane Street, where it’s been incredibly successful internally, and wrote a…
Visit thume.ca → RSS feedOn Hacker News Dec 2016 – Dec 2023
Software Engineering 9Tools & Productivity 7Systems & Performance 4Math & Algorithms 2Life, Learning & Writing 2Hardware & Electronics 2
In this post I’ll attempt the fun stunt of designing a system that could serve the full production load of Twitter with most of the features intact on a single (very powerful) machine. I’ll start by showing off a Rust prototype of the core tweet distribution data structure handling 35x full load by fitting the hot set…
One day a coworker mentioned that he was thinking about APIs for distributed compute clusters and I jokingly responded “clearly the ideal API would be simply calling telefork() and your process wakes up on every machine of the cluster with the return value being the instance ID”. I ended up captivated by this idea: I…
In the course of trying to figure out how to smoothly zoom timelines of a billion trace events, I figured out a cool tree structure that I can’t find elsewhere online, which it turned out two of my friends have independently derived after not finding anything on their own searches. It’s a way of implementing an index…
Edit 2017/12/10: So I screwed up, I thought I was safe confirming it in two different ways but I was using an external monitor and all of the below is accurate only for a specific multi-monitor case. Skip to the bottom to read about my new results.