keynha

Namespaces, cgroups, mounts, and seccomp account for only 3ms of your 30ms cold start, and exec plus runtime init is 22ms. Is that 22 about the floor for full Node compatibility, or is there still room to cut it?

How do you decide when the cache-hit savings are worth riding out a degrading route versus eating the cold start on a healthy one?

Makes sense, thanks. The fixed 10-per-bucket cap is what keeps every viewport the same size no matter the zoom, that's the elegant bit.

This is really nicely done, the log zoom makes 4M events feel navigable instead of like a wall. How are you serving a viewport at that scale: precomputed levels of detail, or a query against the raw events on each pan and zoom?

0.84 Spearman fidelity to the MiniLM teacher at ternary precision is a striking result. How much of that is the quantization-aware training doing the work, versus what a post-training ternary quant of the same encoder would give you?

Dropping the explicit P_DIRTY flag in 1.0 is a neat change. What tracks which pages still need to be written back at commit now that the flag is gone?