• ATS1312@lemmy.dbzer0.com
    link
    fedilink
    arrow-up
    9
    ·
    edit-2
    11 hours ago

    Anyone else convinced AES-256 is not long enough a but length for full disk encryption anymore? Even the 512-bit schemes are just trying some offset-salts that, with ~10TB I worry about providing close-enough statistical significance for cryptanalysis.

    With 100TB drives?! I am less worried and more… convinced its a problem.

    • WhoIzDisIz@lemmy.today
      link
      fedilink
      arrow-up
      1
      ·
      3 hours ago

      At the current rate of development, every city will have its own space heater data center, so with all that computational power available to just any maligned local city government dignitary, you’re right to be concerned.

    • naeap@sopuli.xyz
      link
      fedilink
      arrow-up
      5
      ·
      6 hours ago

      I have to admit, that I never looked into the technical details of full disk encryption

      If I understand you correctly, they are using the same key for all the data and with larger amounts of data statistical analysis becomes feasible
      Did I get this right?

      Couldn’t that be solved by using a root key + salt per block/sector/file/whatever?

      I’d still only need the one root key and with every block the actual encryption key changes

      I was thinking about perfect forward secrecy and that was the first thing, I could come up with

      But, I’m absolutely not a crypto/math guy, so probably I don’t know enough to really add something to the discussion/solution…