Thomas Camlong
dc46057544
Two things this settles.
First, the end-to-end number. Every Phase 2 result so far was an incremental A/B against
the build immediately before it, which cannot be summed. Rebuilt the merge-base
(06724cfeb) from a clean worktree and ran 2+2 alternated against HEAD, reporting cgroup
anonymous memory - the figure that counts against a container limit and cannot be
reclaimed:
stage before after
peak anon 376.6 -> 279.3 MiB (-26%)
boot idle 130.5 -> 62.4 MiB (-52%)
after sign-in 247.4 -> 142.9 MiB (-42%)
dashboard 293.4 -> 170.4 MiB (-42%)
4 tabs 366.5 -> 258.2 MiB (-30%)
after soak 323.7 -> 200.8 MiB (-38%)
peak node RSS 430.7 -> 336.2 MiB (-22%)
No before/after range overlaps at any stage. Costs: CPU 23.5 -> 27.7s (+18%, solid,
non-overlapping). Median board load reads +30% but is NOT established - individual loads
were [506,383,1233] and [848,431,574] before against [780,808,515] and [621,579,1004]
after, heavily overlapping with no cold-then-warm pattern, so six samples per side cannot
separate a real regression from the variance.
compare-stress-runs.mjs now leads with anonMiB, which the harness recorded all along but
the report was ignoring in favour of memory.current - a figure that includes page cache
and swings 60+ MiB between runs of a byte-identical image.
Second, jemalloc. The largest attributed-but-unfixed item is ~38 MiB of V8 zone memory
that musl frees to its own pools and never returns to the kernel, so an allocator that
decays spans back to the OS looked like the answer. Tested on the byte-identical image
with only HOMARR_MALLOC differing, asserting the "Using jemalloc" log line:
boot idle 62.4 -> 80.3 MiB (+29%)
after soak 215.5 -> 279.6 MiB (+30%)
4 tabs 259.9 -> 319.6 MiB (+23%)
CPU 28.0 -> 24.5 s (-13%)
The hypothesis was backwards. jemalloc keeps more freed memory on purpose - thread caches
plus time-based dirty-page decay - and spends it on speed. 13% cheaper CPU, 18-30% more
resident memory. Reverted, and it closes off "swap the allocator" as a RAM direction:
musl is already the frugal choice here.
Live Demo 🎮 • Install 💻 • Translations 🈺 • Discord 👋
- 🖌️ Highly customizable with an extensive drag and drop grid system
- ✨ Integrates seamlessly with your favorite self-hosted applications
- 📌 Easy and fast app management - no YAML involved
- 👤 Detailed and easy to use user management with permissions and groups
- 👥 Support for single sign on via OIDC / LDAP
- 🙊 Safe encryption using BCrypt and AES-256-CBC for your valuable data
- 🕔 Realtime widget updates using WebSockets, tRPC and Redis
- 🔍 Search through thousands of data points in supported integrations or your data in Homarr using the fast built-in search
- 🦞 Icon picker with over 11K icons
- 🚀 Compatible with any major consumer hardware (x86, Raspberry Pi, old laptops, ...) and most OS (Windows, Linux, TrueNAS, Unraid)
- 🖥️ Extensive Kubernetes support with Helm for efficient scaling & high reliability
Please click here for official installation instructions
Homarr is a free-to-use open source project maintained by volunteers and developers from all over the world.
We publish under the Apache License 2.0 license which allows commercial usage.
We invest multiple hours daily in providing support, developing Homarr, integrating to third party software and more.
We also pay for licensing and server hosting fees.
Please consider helping us cover these costs to enable the future development of Homarr. Thank you!
Please click here to sponsor us at OpenCollective
You can also support us by helping with translating the entire project to as many languages as possible or contributing directly to the code or documentation. Please read our Contribution Guidelines. All contributions, regardless of their size or scope, are welcome and highly appreciated! Thank you ❤️
Sponsors
Thanks to your generous sponsors, we can continue to build Homarr. Check them out for high-quality and easy-to-use development tools. Feel free to contact us at homarr-labs@proton.me if you wish to become a sponsor.
