• 0 Posts
  • 16 Comments
Joined 2 years ago
cake
Cake day: June 25th, 2023

help-circle
  • Probably best to look at it as a competitor to a Xeon D system, rather than any full-size server.

    We use a few of the Dell XR4000 at work (https://www.dell.com/en-us/shop/ipovw/poweredge-xr4510c), as they’re small, low power, and able to be mounted in a 2-post comms rack.

    Our CPU of choice there is the Xeon D-2776NT (https://www.intel.com/content/www/us/en/products/sku/226239/intel-xeon-d2776nt-processor-25m-cache-up-to-3-20-ghz/specifications.html), which features 16 cores @ 2.1GHz, 32 PCIe 4.0 lanes, and is rated 117W.

    The ostensibly top of this range 4584PX, also with 16 cores but at double the clock speed, 28 PCIe 5.0 lanes, and 120W seems like it would be a perfectly fine drop-in replacement for that.

    (I will note there is one significant difference that the Xeon does come with a built-in NIC; in this case the 4-port 25Gb “E823-C”, saving you space and PCIe lanes in your system)

    As more PCIe 5.0 expansion options land, I’d expect the need for large quantities of PCIe to diminish somewhat. A 100Gb NIC would only require a x4 port, and even a x8 HBA could push more than 15GB/s. Indeed, if you compare the total possible PCIe throughput of those CPUs, 32x 4.0 is ~63GB/s, while 28x 5.0 gets you ~110GB/s.

    Unfortunately, we’re now at the mercy of what server designs these wind up in. I have to say though, I fully expect it is going to be smaller designs marketed as “edge” compute, like that Dell system.



  • It can be a one-time setup.

    Right up until your laptop gets its motherboard replaced and won’t boot due to a MOK-signed module (in my case it was ZFS, which I needed for the machine to actually function).

    At which point you

    • Switch secure boot from enforcing to permissive mode (note you can’t turn it off entirely, or the enrollment will fail with an error that your system doesn’t support secure boot).
    • Boot into your OS.
    • Find the arcane command to re-enroll the MOK. That’s sudo mokutil --import /var/lib/shim-signed/mok/MOK.der (for Ubuntu derivatives and probably others), in case someone finds this post in the future.
    • Reboot again, accept enrolling the key.
    • Reboot again, and switch back to enforcing.

    If you have a BIOS password, encrypted filesystem, and all the other moving parts that make having secure boot enabled actually a meaningful exercise, this is neither a fun, nor particularly quick process.

    As for modules being signed automatically when built by DKMS, I’ve never had an issue with that.





  • It looks like it goes away (I went into the settings for one contact, where there’s a toggle to switch back to SMS/MMS).

    Unfortunately, I can’t figure out how to change it back to RCS for that conversation now.

    Also, on a more personal note: the baby blue background is ugly AF. Like actually just heinous. I’d rather it wasn’t there, you get a perfectly good indicator RIGHT ABOVE THE TEXT INPUT FIELD what it’s about to send.


  • I recently bought a Boox Palma, which is a phone-size Android device with a real E-Ink display.

    It’s not a phone (WiFi/Bluetooth only, no mobile radio), and with 4-bit greyscale it’s definitely an adjustment to use with a lot of apps (it has per-app DPI & contrast controls to help), but they’ve done a lot of work on the refresh rate to make it feel responsive.

    It even has midrange-phone specs (SD 6xx series CPU, 6GB RAM, 4Ah battery), with full Google Play, so it’s a quite usable Android device overall. Like most modern E-Ink devices, has a CCT warm-to-cool frontlight, so great for night-time use.

    Now would I want to use it as my only, everyday device (if it was a phone too)? Probably not. Could I? Almost certainly.

    Colour E-Ink is still quite limited (in contrast, and resolution), but I expect the patents on that are quite a bit newer and we won’t be seeing so much movement in that area so soon.


  • To expand on @doeknius_gloek’s comment, those categories usually directly correlate to a range of DWPD (endurance) figures. I’m most familiar with buying servers from Dell, but other brands are pretty similar.

    Usually, the split is something like this:

    • Read-intensive (RI): 0.8 - 1.2 DWPD (commonly used for file servers and the likes, where data is relatively static)
    • Mixed-use (MU): 3 - 5 DWPD (normal for databases or cache servers, where data is changing relatively frequently)
    • Write-intensive (WI): ≥10 DPWD (for massive databases, heavily-used write cache devices like ZFS ZIL/SLOG devices, that sort of thing)

    (Consumer SSDs frequently have endurances only in the 0.1 - 0.3 DWPD range for comparison, and I’ve seen as low as 0.05)

    You’ll also find these tiers roughly line up with the SSDs that expose different capacities while having the same amount of flash inside; where a consumer drive would be 512GB, an enterprise RI would be 480GB, and a MU/WI only 400GB. Similarly 1TB/960GB/800GB, 2TB/1.92TB/1.6TB, etc.

    If you only get a TBW figure, just divide by the capacity and the length of the warranty. For instance a 1.92TB 1DWPD with 5y warranty might list 3.5PBW.





  • qupada@kbin.socialtoGaming@lemmy.zipHold my beer - Bungie
    link
    fedilink
    arrow-up
    37
    arrow-down
    1
    ·
    1 year ago

    Plays out in small tech companies too, albeit in a slightly different way.

    Got that carrot dangled in front of me at a past job. Company was past start-up phase; self-supporting and doing ok, but not outrageously well. Promises of riches should the company be “noticed” and bought for an outrageous amount.

    Of course none of that accounted for the CEO (founder and 85% shareholder) being an absolute crazy person, who would change the development roadmap into making a vastly different product than the one we (the techies) believed in, TURN DOWN THE OUTRAGEOUS SUM BECAUSE HE THOUGHT HE COULD GET A BETTER OFFER, basically run the company into the ground, and wind up selling it for a pittance (which would have made the employees’ share a pittance of a pittance).

    I mean most of us had already left by that point, but finding out around 4 years after that he’d turned down about $150M and wound up selling out for $3M, that stung a little.