• 0 Posts
  • 33 Comments
Joined 1 year ago
cake
Cake day: July 9th, 2023

help-circle














  • Maybe yeah. Also got the sense from the strong opinions that this is a preexisting debate, presumably in the context of continuous workloads or cached arrays with minimal spindown intervals. In that context it’s true that rotational disks still often win in energy efficiency and robustness (assuming we’re comparing them to consumer SSDs and not the latest enterprise u.2 stuff that’s rated for continuous work).


  • Not sure what everyone is arguing about here. Clearly SSD is better for intermittent r/w, whereas HDD can be more efficient at continuous r/w (especially in terms of watts/TB)

    Just looking at specs should be enough to see that. SSDs can idle in ready state at close to 0 draw (~0.05w) whereas HDD requires continued rotation to remain ready. So consider an extreme case of writing for 1 minute then maintaining ready state for the rest of the day. For that the SSD will be far more efficient, obviously.


  • True, and interesting since this can be used as a statistical lever to ignore the exponential scaling effect of conditional probability, with a minor catch.

    Lemma: Compartmentalization can reduce, even eliminate, chance of exposure introduced by conspirators.

    Proof: First, we fix a mean probability p of success (avoiding accidental/deliberate exposure) by any privy to the plot.

    Next, we fix some frequency k1, k2, … , kn of potential exposure events by each conspirators 1, …, n over time t and express the mean frequency as k.

    Then for n conspirators we can express the overall probability of success as

    1 ⋅ ptk~1~ ⋅ ptk~2~ ⋅ … ⋅ ptk~n~ = pntk

    Full compartmentalization reduces n to 1, leaving us with a function of time only ptk. ∎

    Theorem: While it is possible that there exist past or present conspiracies w.h.p. of never being exposed:

    1. they involve a fairly high mortality rate of 100%, and
    2. they aren’t conspiracies in the first place.

    Proof: The lemma holds with the following catch.

    (P1) ptk is still exponential over time t unless the sole conspirator, upon setting a plot in motion w.p. pt~1~k = pk, is eliminated from the function such that pk is the final (constant) probability.

    (P2) For n = 1, this is really more a plot by an individual rather than a proper “conspiracy,” since no individual conspires with another. ∎