• GameGod@beehaw.org
    link
    fedilink
    English
    arrow-up
    59
    ·
    1 year ago

    We already had this, it’s called Intel Optane Persistent Memory and Intel killed it off last year: https://www.intel.com/content/www/us/en/products/docs/memory-storage/optane-persistent-memory/overview.html

    The memory speed was slightly slower than DDR4 but the benefits didn’t seem to outweigh the downsides. I think it probably kicked a lot of ass for specific use cases (eg. in-memory database that needs persistence), but the market was too small. Plus, SSDs are getting so ridiculously fast that it would put pressure on a product like this too.

  • wrath-sedan@kbin.social
    link
    fedilink
    arrow-up
    31
    ·
    edit-2
    1 year ago

    Can’t wait for my ram to last 1000 years just for the hinge on my laptop screen to last 2 (guess what just broke on my laptop after 2 years)

  • BuxtonWater@beehaw.org
    link
    fedilink
    English
    arrow-up
    24
    ·
    1 year ago

    Perfect for when civilization collapses and we have to do some wasteland 2 shenannigans to get the lost knowledge of the past back by hoarding laptops.

  • Catsrules@lemmy.ml
    link
    fedilink
    arrow-up
    24
    arrow-down
    1
    ·
    1 year ago

    Of course, this is still a new and emerging technology and it’s too early to say when we might see it in our devices, or how much it will cost.

    Looks really cool, buy yeah my guess is i will cost to much to be viable for most things.

    • weew@lemmy.ca
      link
      fedilink
      arrow-up
      2
      ·
      edit-2
      1 year ago

      it’ll hit business servers first… speed and power draw = profits. Even if they cost 1000x more than SSDs, the power savings and speed alone could pay for itself in a datacenter.

  • brie@beehaw.org
    link
    fedilink
    English
    arrow-up
    15
    ·
    1 year ago

    This sounds neat, but it also seems like it does not have much practical advantage over hibernation except faster wake.

    • weew@lemmy.ca
      link
      fedilink
      arrow-up
      21
      arrow-down
      1
      ·
      edit-2
      1 year ago

      The power draw and nonvolatility could mean it can replace SSDs and hard drives entirely. Just store everything in RAM.

      • CanadaPlus@lemmy.sdf.org
        link
        fedilink
        arrow-up
        7
        ·
        1 year ago

        So you know those encrypted files you might have? You know how your computer is able to display them? Yeah, you enter the passcode and then it and any information derived from it is (probably, barring special hardware) stored in RAM to in order do that. This isn’t usually an issue because turning your computer off will quickly erase the RAM and it can no longer be used to open that file.

        • XTornado@lemmy.ml
          link
          fedilink
          arrow-up
          6
          ·
          edit-2
          1 year ago

          I mean… this already happens now if you leave the PC open or in sleep, for example with bitlocker.

          And I guess in this case it could allow when shutdown if the memory has not been cleared… But most of the time that is not a problem, most people aren’t attacked by pulling the info directly from the hardware.

          And in case that is a possible attack, I expect a mechanism of the app(s) to clear that region on shutdown or whatever is done or even better maybe the OS would do it as part of the existing memory protection mechanisms. I mean most apps already clear the memory as soon as possible if it is not needed to avoid having secrets or unencrypted information on memory.

          • CanadaPlus@lemmy.sdf.org
            link
            fedilink
            arrow-up
            3
            ·
            1 year ago

            I didn’t actually look more closely at how this works, but some kinds of memory are hard to safely erase. If I had to use this, I’d make sure there was a secure wipe of the whole thing minus whatever the system itself is using at shutdown.

            It wouldn’t completely break security, if that even makes sense as a concept, but it wouldn’t help. Hardware security is the strongest kind.

                • ricecake@beehaw.org
                  link
                  fedilink
                  arrow-up
                  3
                  ·
                  1 year ago

                  This is already a thing we need to deal with, security wise. An application making use of encryption doesn’t know the condition of what it views as ram, and it could very well be transferred to a durable medium due to memory pressure. Same thing with hibernation as opposed to suspension.

                  Depending on your application and how sensitive it is, there are different steps you can take to deal with stuff like that.

    • Lupec@lemm.ee
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      Yeah, sounds like a best of both worlds situation where you wouldn’t have to decide whether to suspend or hibernate anymore

    • FaceDeer@kbin.social
      link
      fedilink
      arrow-up
      15
      arrow-down
      1
      ·
      1 year ago

      Sounds more like a very small security risk, to me. For most people if someone steals their laptop it doesn’t really matter what kind of RAM is in it.

      If you’re in some kind of high-security role then use a laptop with volatile RAM instead. Non-volatile will probably be more expensive than the old stuff for quite a while so it’s not going anywhere.

      • InsurgentRat@beehaw.org
        link
        fedilink
        arrow-up
        10
        ·
        1 year ago

        Maybe, it depends how it works.

        Memory is often unencrypted and/or contains encryption keys. Many programs rely on the assumption that it’s cleared on powerdown for security.

        Depending on how this memory enters the long term state it seems that a lot of legacy software might become vulnerable to a really simple attack.

        Pulling the plug might no longer be something that forces someone to engage in rubber hose analysis.

        • FaceDeer@kbin.social
          link
          fedilink
          arrow-up
          4
          arrow-down
          1
          ·
          1 year ago

          Many programs rely on the assumption that it’s cleared on powerdown for security.

          In a world with ULTRARAM those programs will need to be rewritten to operate under new assumptions.

          • InsurgentRat@beehaw.org
            link
            fedilink
            arrow-up
            7
            ·
            1 year ago

            Sure of course of course but umm have you seen software?

            There are still windows xp computers on the internet.

            It’s not insurmountable, and of course I have no idea if/how this will roll out.

            Just it seems to mess with a rather deep assumption we have about how computers operate when we develop software and threat models.

            • bedrooms@kbin.social
              link
              fedilink
              arrow-up
              5
              ·
              1 year ago

              The better example is that there are still programs compiled in Win 95 running here and there if I’m correct.

            • FaceDeer@kbin.social
              link
              fedilink
              arrow-up
              1
              ·
              1 year ago

              Sure, but anyone who’s still running WindowsXP on their laptop wants to suffer.

              As always, the security-minded will be able to secure their devices and everyone else will not care until it bites them. ULTRARAM isn’t revolutionary in that regard.

      • InsurgentRat@beehaw.org
        link
        fedilink
        arrow-up
        6
        ·
        edit-2
        1 year ago

        I admit to being out of the game for a while but how common is RAM encryption?

        wouldn’t the overhead violate half the point of RAM?

        • d3Xt3r@lemmy.nz
          link
          fedilink
          arrow-up
          3
          ·
          edit-2
          1 year ago

          There’s always a tradeoff in computing between security v/s performance/overhead, so the value of it depends on your threat model, and the attack vectors you’re expecting.

          Anyways, RAM encryption is generally “available” in various forms, depending on the mobo, CPU, software used etc, but it’s not commonly enabled/used. Most AMD boards (at least, mid-range and above should) have an option in the BIOS to enable Secure Memory Encryption (SME). This allows the OS to selectively encrypt memory pages, making use of a hardware AES engine that sits outside of the CPU.

          There’s also Transparent SME (TSME), which encrypts the entire memory and works completely independent of the OS and software. Usually only high-end/workstation boards have this, and it also requires a Ryzen PRO CPU. TSME also has a much lower overhead, I recall reading somewhere it’s something like only 5%.

          I believe Intel also has something similar, but I never looked into it.

          AMD have a whitepaper available with an overview on how this stuff works, if you’re interested: https://www.amd.com/system/files/TechDocs/memory-encryption-white-paper.pdf

  • Bizarroland@kbin.social
    link
    fedilink
    arrow-up
    7
    ·
    edit-2
    1 year ago

    I have to regretfully say I would have had an apocalypse bingo but I didn’t have “Laptop of Dorian Gray” on my scorecard