Performance and scaling consultant who analyzes the full stack - frontend, backend and infrastructure - to generate more revenue and cut costs for businesses.
Remote • Available for monthly retainer and bespoke projects.
I was recently adding a bunch of files to a new Git repo, but I couldn't be sure if I was about to commit a 100MB file. The following bash command should print out a sorted list of unstaged files and their size, so you can check if
I recently managed to get a 4iiii Precision left-only crank power meter. I fitted it and did a 40 minute indoor session. I have a dumb trainer, but I've found the power curve online. It was showing that I could expect ~200W for 30km/h on a certain
I spent most of 2017 and 2018 running with little structure or knowledge of pacing. When I first started, I kept trying to PB my 5km time literally every single run. A lot of my time was spent injured from ramping up too quickly, and running too quickly. I bought
As of October 31st 2019, my current running PBs are the following: * 5km: 19:49 (3:58 min/km - November 2018) * 10km: 42:13 (4:13 min/km - October 2019) * Half marathon: 1:30:17 (4:17 min/km - October 2019) * Marathon (unofficial): 3:39:46 (5:
Last year I wrote a year-in-review for my twenty-third [https://daniellockyer.com/twenty-three] birthday. And now it's time for another! In the last year, I have: * visited Chiang Mai - Here for my birthday last year, until just after NYD. Came back here during summer to regain fitness
If you read my blog or follow me on Twitter you may have noticed I'm a huge fan of flamegraphs. I've previously written about how I use them to profile PHP and Node.js applications, and I help maintain rbspy, a Rust profiler for Ruby, which
Last October I made Upgrade your PHP [https://upgradeyourphp.com/], a site to let people know that PHP 5.6 and PHP 7.0 were about to reach their End of Life, and would no longer be officially supported. In the time since, I've been helping people upgrade
This post is a case study of how I profiled a Node.js application, and how using flamegraphs allowed me to make the boot time of Ghost 20% faster. I remember trying to profile Node.js many years ago and it was much harder than it is now. When researching
Earlier today I was looking at a couple of sites to see why they were much slower than I expected. I ran them through GTmetrix and saw the following in the waterfall tab. Straightaway I could see that this is the classic pattern produced by trying to load a bunch
I recently added support for Ruby 2.5.4, 2.5.5 and 2.6.0-2.6.3 to rbspy, the sampling profiler written in Rust. I have replaced the original Flamegraph library by Brendan Gregg with Inferno, a Rust port. I also updated and merged a PR for handling