How we took our warm start service from 2,099 req/s on Node + SQLite to 10,700 req/s on a compiled Bun binary, and found a subtle memory leak that only exists in Bun's HTTP model.
“We rewrote the code to be much more performant as part of a migration to Bun, and achieved a 5x throughput increase!” Okay but to bill this as a difference in runtimes, you have to compare the optimized code on both runtimes, rather than the unoptimized code on node to the optimized code on bun? Bun’s dev team was annoying before but now that they’ve been bought by anthropic they’re truly insufferable.
“We rewrote the code to be much more performant as part of a migration to Bun, and achieved a 5x throughput increase!” Okay but to bill this as a difference in runtimes, you have to compare the optimized code on both runtimes, rather than the unoptimized code on node to the optimized code on bun? Bun’s dev team was annoying before but now that they’ve been bought by anthropic they’re truly insufferable.