Memory

The Kernel Tax: IPC Latency and the Cost of Boundary Crossings

Every pipe round-trip crosses the user/kernel boundary four times. Each crossing costs a minimum of 101 ns on modern hardware. Shared memory crosses it zero times. A forensic breakdown of what the kernel actually charges per message, and why Tachyon reaches 56.5 ns where pipes cost 6.2 µs.

The Invisible Lock: Cache Coherency and the Physics of False Sharing

You removed the mutex. The CPU added a hardware lock. A deep dive into how the MESI protocol and false sharing silently destroy multi-core scaling.

The O(1) Illusion: Why Pointer Chasing is the Death of Throughput

Algorithmic complexity assumes memory is flat and fast. It isn’t. A deep dive into why contiguous arrays destroy linked lists on modern superscalar CPUs.