New 12TB=96Tb HD. Non-recoverable error rate: 1e-15 per bit. E[complete drive reads per error]: 10 https://www.wdc.com/content/dam/wdc/website/downloadable_assets/eng/spec_data_sheet/2879-800074.pdf …pic.twitter.com/W3ztIYiudi
Show this thread
Want to write to your filesystem safely? No worries, the manpage has all the latest rumors http://danluu.com/file-consistency/ …pic.twitter.com/uhGU1LAM2w
-
-
-
In the fine print, they say their annualized failure rate calculation is based on a workload of 219TB/yr for a 12TB drive
Show this thread -
It's a little concerning that the APFS devs say data checksums are unneeded because HDs are reliable, contradicting every study on HD errorspic.twitter.com/uHNdsIwhXp
Show this thread
End of conversation
New conversation -
-
-
What a strange response. This is exactly what it looks like when programs have crash consistency bugs https://github.com/prometheus/prometheus/issues/1967 …pic.twitter.com/2p0907ozxU
Show this thread -
It's like watching people using mongo respond to aphyr: "there can't be bugs because I only rarely hear about data corruption".
Show this thread - 2 more replies
New conversation -
-
-
@danluu Really enjoyed this paper about filesystems. Thank you for writing it. Any chance to get ZFS added to the table of filesystems? -
@mihalis68 I would love to have the time to reproduce the results on current filesystems, including ZFS. Not likely soon, though.
End of conversation
New conversation -
-
-
you can tune a filesystem...
-
But you can't file a tuning system?
- 2 more replies
New conversation -
-
-
Want to write to your filesystem safely? Use SQLite.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.