GitHub - DataDog/go-profiler-notes: felixge's notes on the various go profiling methods that are available.
felixge's notes on the various go profiling methods that are available. - DataDog/go-profiler-notes
Profiling to Reduce Memory Allocations
Using pprof and go tool trace to find and fix a 100x memory allocation issue caused by compression middleware.
Hey linker, can you spare a meg?
Tailscale on iOS runs as a special kind of app, a Network Extension. This lets us run in the background, so we can secure traffic from all of your applications, without them having to change anything. But with this power comes a memory straightjacket. Normal iOS apps can use 5GB or so of memory before iOS kills them. We get 15MB. With an “M”.
Running a million-board chess MMO in a single process · eieio.games
How one million chessboards works
Running a million-board chess MMO in a single process · eieio.games
How one million chessboards works
Profiling gin with pprof
Go comes with great tools for profiling out of the box. It’s one of the features I’ve come to love. I’m not going to go into detail about pprof, but if you need a primer, Julia Evans has a great post about it on her blog. Instead, I’ll try and show a real world example of the optimizations you can make using pprof. I’ll use pprof on mouthful, a gin based server, to see what I could do to make it just a bit faster. It turns out I can detect a rather stupid mistake I’ve made.