5 ms·

I'm pretty sure whatever that community experienced is more anecdotal that statistically provable...

• more_corn 1y ago
  I’ve worked at a scale that is statistically relevant. Tens of thousands of drives under my control. I’ve seen a ton of different failure modes. Some of our anecdotes are actually useful. The problem with book and lab theory is that sometimes the theoretical problems don’t manifest (SSD wear out for example) and sometimes the minor seeming things turn out to matter a lot.