Lambda isn't leaking memory, your metrics are lying to you
engineering.taktile.com
Lambda isn't leaking memory, your metrics are lying to you
1–6 of 6 posts
Re: Lambda isn't leaking memory, your metrics are lying to you
#2Claude write me a post-mortem. Make no mistakes. Add a big image of a hero banner instead of rendering it in HTML for some reason.
I use AI myself (it's essentially non-optional at work right now), and it's not like this is useless information, but god, I'm just so saturated of this writing style.
Re: Lambda isn't leaking memory, your metrics are lying to you
#3I am wondering about the last part of glibc malloc. Isnt that exactly the problem reported for the last 10+ years with glibc? And the common solution is to use tcmalloc or jemalloc which hasnt these problems?
Re: Lambda isn't leaking memory, your metrics are lying to you
#4Claude write me a post-mortem. Make no mistakes. Add a big image of a hero banner instead of rendering it in HTML for some reason. I use AI myself (it's essentially non-optional at work right now), and it's not like this is useless information, but god, I'm just so saturated of this writing style.
> I'm just so saturated of this writing style.
I saw at least one tense error, so much for the idea of intentionally introducing grammar mistakes to demonstrate personhood.
Re: Lambda isn't leaking memory, your metrics are lying to you
#5Claude write me a post-mortem. Make no mistakes. Add a big image of a hero banner instead of rendering it in HTML for some reason. I use AI myself (it's essentially non-optional at work right now), and it's not like this is useless information, but god, I'm just so saturated of this writing style.
Dammit. For a long time, I was blissfully clueless about AI writing style. But I recently read something that screamed en-SLOP, and now I can't unsee it anywhere. It hit me immediately with this article.
I want to go back to being dumb and naive. Give me the blue pill, please!
Re: Lambda isn't leaking memory, your metrics are lying to you
#6"RSS lies. Your process might not be using that memory. The allocator might be hoarding it."
Interesting writeup, but:
No, anonymous AI author. Your process is using that memory, for its allocator. Features like lower-latency allocations don't come for free, even when they turn out severely suboptimal for your particular case. Your code isn't using that memory, but a support library is. It is very much in your process.