News for hackers. Stuff that matters.
Re:Nanolog supports logging with 7 ns median latency by kolbe on Friday August 23, 2024 @11:05AM (#41329624)
I could swear I did a deep dive into Spdlog vs Nanolog six months ago, and the performance differences weren't nearly this stark
Re:Nanolog supports logging with 7 ns median latency by synergy20 on Friday August 23, 2024 @12:23PM (#41330330)
what do you mean? considering spdlog is the de facto logger for c++
Re:Nanolog supports logging with 7 ns median latency by kolbe on Friday August 23, 2024 @03:15PM (#41332135)
Nano is claiming insanely better performance over spdlog, which confuses me
Re:Nanolog supports logging with 7 ns median latency by odygrd on Monday August 26, 2024 @01:11PM (#41359315)
spdlog is designed as a general purpose logging library and it can’t beat low latency loggers. It doesn’t scale for multiple threads because it’s async mode is using a mutex and a cv to notify the background thread.

You can find some logging libraries benchmarks here https://github.com/odygrd/quill?tab=readme-ov-file#-performa...

Re:Nanolog supports logging with 7 ns median latency by linhns on Saturday August 24, 2024 @01:46AM (#41335940)
I believe the performance reported in their paper is circumstantial. It’s not that much faster when I tried it, and not worth the horrible macro syntax.