3 ms·
Dunno know I've ever heard conventional wisdom that said NAND flash was particularly short lived on the shelf vs other options. E.g. never heard of mass deaths
by zamadatix 2y ago
Dunno know I've ever heard conventional wisdom that said NAND flash was particularly short lived on the shelf vs other options. E.g. never heard of mass deaths of files on USB drives from sitting in drawer for a year or two, usually just typical endurance/usage deaths from cheap flash.
Hell, I just took my very first SSD (60 GB OCZ Vertex 2) and booted the old install fine after it had sat on a shelf for ~7 years. Of course that just had the Windows boot files on NTFS so I can't say for sure no bitrot had occurred. In the same bin/nostalgia period I had an old 2 GB USB 2.0 flash drive with Ubuntu 9.10 on it that did validate the fs image fine from the same bin as well.
I remember https://www.youtube.com/watch?v=igJK5YDb73w https://www.youtube.com/watch?v=igJK5YDb73w this guy was doing some intentional tests with modern cheapo drives but it hasn't even reached time for the year 2 check quite yet.
- 0cf8612b2e1e 2y agoPDF[0], but this industry presentation said that for a consumer drive kept at 25C, one should expect 58 weeks of unpowered data storage (page 27). Significantly worse for higher temps. I appreciate those are probably bare minimum threshold numbers which have probably changed, but that makes me uncomfortable vs a HDD where I have not heard such a qualifier. Just given the cost per TB, I will continue to use HDDs for offline backups. [0] https://www.jedec.org/sites/default/files/Alvin_Cox%20%5bCompatibility%20Mode%5d_0.pdf https://www.jedec.org/sites/default/files/Alvin_Cox%20%5bCom...
- zamadatix 2y agoSince at least 2010 there are flash standards which define a floor and no such standard at all for HDD? Choose what you want of course but I'll take a 15 year old minimal standards guarantee over no standards guarantee at all.