Nixpkgs evaluation time share #2

Open
opened 2026-02-24 12:34:11 +00:00 by puck · 0 comments

The performance benchmarks compare relatively lean NixOS system closures (a singular service, plus dependencies) asd show relatively small time difference compared to the difference implied on a larger closures. It would maybe be nice to see the static overhead of evaluating nixpkgs mentioned, since, from what I know (and, of course, this may be wrong) that would be a large part of evaluating e.g. grafana's systemd unit file, and would likely be a similar share of performance on bath the partial and full closure evaluations, at least in the examples in the document.

The performance benchmarks compare relatively lean NixOS system closures (a singular service, plus dependencies) asd show relatively small time difference compared to the difference implied on a larger closures. It would maybe be nice to see the static overhead of evaluating nixpkgs mentioned, since, from what I know (and, of course, this may be wrong) that would be a large part of evaluating e.g. grafana's systemd unit file, and would likely be a similar share of performance on bath the partial and full closure evaluations, at least in the examples in the document.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
qyriad/rfd-modular-dynamicism#2
No description provided.