Both boxes run the same OS stack (Debian-based NAS distro, Docker), so the only variables are the drive and the switch port.

Box A — 2011 laptop: Celeron B815 (2C/2T, 1.6 GHz), 4 GB DDR3, Realtek 100 Mbps, 256 GB SATA SSD + 320 GB 5400 rpm in the optical bay. Box B — 2011 mini PC: i5-2520M, 8 GB DDR3, gigabit, 128 GB SSD + 3 TB WD Red Plus.

Same library, same client:

  • Immich first load: 24 s on the 5400 rpm 2.5-inch, 0.25 s after moving the DB + UI to the 256 GB SSD. ~95x, on identical hardware.
  • SMB to Box A: 11.4 MB/s up, 11.3 MB/s down (85-91 Mbps — the link is saturated, the disk is not the limit).
  • hdparm -t on the SATA SSD: 506 MB/s.
  • WD Red Plus sequential write: 107 MB/s. 5400 rpm 2.5-inch buffered read: 88 MB/s.
  • Box B idle 20 W, under sustained load ~40 W.

The takeaway that surprised me: the photo blobs are fine on spinning rust. What kills perceived speed is putting the Postgres DB and the thumbnail cache on a 5400 rpm drive — every page load turns into a scatter of small random reads. Move those two, leave the originals where they are.

The other half is the network. Once the SSD was in, the 100 Mbps port became the bottleneck for everything except thumbnail generation, which is pure CPU (about 91 thumbnails a minute, so a 12k library is roughly two hours of solid work).

I keep the full numbers, and the method used to get them, here: https://lowspechq.com/best/cheap-ssd-for-nas/ Source for the site builder if you want to check the claims against the code: https://github.com/lowspecai/lowspec-ai

Disclosure: I run lowspechq.com. That page carries one clearly labelled AliExpress affiliate link (an SSD listing) and nothing else; every number above was measured before any link existed.

  • curbstickle_lw@lemmy.worldMEnglish
    0·
    3 days ago

    The first comment here from you was llm generated and referenced your site, followed by this post referencing your site.

    This comes across like self-promotion, which has rules. Since the post has a lot of comments at the moment I’m going to lock it, but if you keep only posting/commenting things about your site without otherwise participating then they will be removed going forward.

    Edit: Revising that. Half are noting this as blogspam. This is being removed.

  • Ebby@lemmy.ssba.comEnglish
    0·
    3 days ago

    I remember running a website off a RAM disk for a little while to maximize page speed and game servers. Laptop drives were slower than 5400 and it did give the site a good boost.

    Very unstable system though. And so much fun.

    • JustEnoughDucks@feddit.nlEnglish
      0·
      3 days ago

      Red plus should be safe from that. The drives they are talking about are bottom of the barrel external HDDs.

      The standard nternal WD blues were used as an actual 5400rpm reference.

      Red plus’s are the helium-fille (higher drives)d upgrade for reds, so they would general be safe too.

      I should measure my 2 4TB and 1 10TB red plus. Might be too quiet to actually get a good reading though. Those things spinning are crazy quiet.

      • micgerne@lemmy.dbzer0.comEnglish
        0·
        3 days ago

        Did you read the article? I’m guessing not, because you’d have read this part

        … and a “5400 RPM Class” Western Digital Red—which, as tested by audio spectral analysis, actually spins at 7,200rpm

        or followed the link to the original investigation and seen this part:

        Known “5400RPM Class”=7200RPM drives: …

        WD80EZAZ-11TDBA0 (WD MyBook 8TB)

        WD80EFAX (WD Red Plus 8TB)

        WD100EFAX/WD101EFAX (WD Red Plus 10TB)

        [WD80EFZX? (old WD Red 8TB)]

  • kalman@dboft.com
    0·
    4 days ago

    Guy put Postgres on a 5400 rpm laptop drive, wondered why Immich took 24 seconds to load, then “discovered” a $20 SSD fixes it — 95x faster, stop the presses. The rest is him learning that a 100 Mbps Realtek port from 2011 is slow and a dual-core Celeron is bad at math.