From: nospam@needed.invalid   
      
   On Sun, 5/18/2025 12:44 PM, Carlos E. R. wrote:   
   ) Unconditional use. Transfer curve is relatively smooth.   
   >   
   > I had not seen this idea of a transfer curve before. Interesting.   
      
   > As long as the computer was not busy those instants. The test can take many   
   hours to run.   
   >   
      
   The benchmarks take tiny samples. The setting for the picture was   
   "1000 samples of 10MB each". That means a kind of crude statistical   
   sample. if you play with the settings, you might notice the interaction   
   between those settings choices, and the amount of "artifacts" in the   
   trace.   
      
   I'm only reading a small fraction of the drive surface. I'm taking   
   a thousand samples. I hope I hit the track that has the huge   
   number of reallocations on it, to get a representative sample.   
      
   I *have* set up a disk before, to bench the whole thing, inch by inch,   
   from end to end. That took five hours and a "custom technique" not   
   suited for others.   
      
   Any scheme you offer for "vetting" disks, can't particularly   
   have a long execution time, as then users won't use it. As long   
   as the bench runs in a couple minutes, most people can manage that.   
      
   The thing is, we need to teach people of the need to "vet" disks   
   before it is TOO LATE. I hate listening to someone whine about   
   their drive full of CRC errors, and their fervent hope all the   
   data will be rescued by some miracle. I hope that maybe, maybe,   
   just once, someone will follow the instructions to bench a drive,   
   and notice it is sick, and get the data off before the disk is ruined.   
      
   It should be noted, that the zone recording scheme of disks,   
   has "peculiar behavior". An ex-employee at a disk company, was   
   explaining some of this on his web site (until the company lawyers   
   detected the leak and shut him down). Some of the disk drives   
   you buy *cannot* have smooth edges in the graph. The ripple in   
   the transfer rate, is due to how the tracks are set up, and   
   the rate on each track can be custom.   
      
   Some drives, just the main zones are visible. Each zone is "flat as   
   a straight edge" on top. For those drives, excursions in   
   storage performance show up well. The hard drives (even modern   
   ones) with "gravel on the edges of the graph", it is then   
   harder to spot real/mechanical trouble as a result. The drive   
   looks "slightly flaky" from the first day you use it. (And no,   
   that is not supposed to be an SMR drive either, it's a PMR   
   with gravel on the bench graph.)   
      
   If at first, your two minute bench does not look "pretty",   
   try adjusting the number of samples and the sample width, and   
   see if that modifies the artifacts from the benching method.   
      
   When hard drives leave the factory, they already have reallocations   
   on them. The reason the "Reallocated" SMART parameter is not an   
   honest, linear, indicator is because customers would "cherry pick"   
   drives and keep sending hard drives back to Newegg, until   
   they got a "perfect one". To stop that from happening, the   
   Reallocated statistic always reads 0 when the drive leaves   
   the factory. This prevents those "special" customers from using   
   a precision Reallocated statistic, to cherry pick drives.   
      
   But because the Reallocated statistic is not an honest one,   
   we cannot "chart" the health of the drive over its lifetime,   
   and plot "reallocations versus time". It is for this reason,   
   that I use the read benchmark as a "proxy for surface damage".   
      
   if the disk drive company won't be honest with us, we have to   
   come up with some sort of solution for an early warning.   
      
      Paul   
      
   --- SoupGate-DOS v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)