How Docker Actually Works: I Rebuilt a Container From Scratch With Linux Commands
No Docker, just Linux: build a root filesystem with debootstrap, isolate it with unshare, wire it to the internet with a veth pair and cap it with cgroups.
// cloudmash.blog
Long-form companions to the Cloudmash videos. Each post follows one request, event or byte through a real system and shows what every component does with it.
No Docker, just Linux: build a root filesystem with debootstrap, isolate it with unshare, wire it to the internet with a veth pair and cap it with cgroups.
A 41 MB Ubuntu image runs a whole Linux on your laptop, and you can start hundreds of them. Are you running 100 Ubuntus? And why is it so small?
Why decoding is bandwidth-bound, how draft-then-verify gives the same answer 3x faster, and how EAGLE-1 to EAGLE-4 learned to guess. With Qwen3.8-27B numbers.
Follow one prompt through llama.cpp on an RTX 5090: how a 16.5 GB Qwen3.8-27B GGUF is memory mapped, copied to VRAM, cached and turned into tokens.
llama.cpp's --moe-cache-mib keeps MoE experts in RAM and caches the hot ones in VRAM: a 94 GB model on a 24 GB RTX 4090, with each MoE flag explained.
How did AI go from a perceptron that couldn't learn XOR to models that turn a sentence into video? Walk the whole road, and find the one idea under it all.
One formula explains why some AI runs on a $2 chip and some needs racks of GPUs. A 4-tier map of model size, memory, and hardware, with real numbers.
What really happens when you run kubectl apply? Follow one Deployment through the API server, etcd, watches, informers, controllers, scheduler and kubelet.
We use cookies to improve this site. Privacy policy