Over the past year or so, I have had an alarmingly high number of USB flash drives fail into "read only" mode. Something like six or more. These varied from cheap Chinese eBay stuff to name brands pretty much equally. So, I got to thinking: How many have failed on me over the last decade or so. Practically none that I can recall. What has changed in manufacture or design that might account for this or is it just coincidental.
I did a search using Startpage, and Duck Duck Go, and didn't find anything that might validate my observations. Please tell me, am I imagining this or is it a real phenomenon? Have any of you noticed increased failure rates of USB flash drives.
There is a motivation to try to get users to migrate from external storage to the cloud. I'm not comfortable with that. I'm strictly VFR. No clouds, low, and slow.
Thanks for any insights you might choose to offer.
(Score: 4, Insightful) by martyb on Thursday January 31 2019, @03:30AM (3 children)
Technology has advanced so that more cells can be constructed out of the same-sized piece of silicon -- the feature sizes get smaller. (Think of how processors have shrunk their feature sizes down from microns (80286) to what's the best there days? 14nm? 12nm?) Fewer atoms are needed to represent a cell.
Also, more data is now stored in each cell (1, 2, 3, or even 4 bits per cell[*])
With less redundancy to store a given data value, things are more susceptible to failure.
[*] See the Wikipedia entry on multi-level cell [wikipedia.org] for an excellent explanation.
Wit is intellect, dancing.
(Score: 5, Funny) by The Mighty Buzzard on Thursday January 31 2019, @05:15AM (1 child)
Plus when you use fewer atoms to store the same amount of data, the atoms wear out faster and you end up with atom dust instead of storage media.
My rights don't end where your fear begins.
(Score: 2) by RS3 on Thursday January 31 2019, @05:24PM
Yeah that quark powder is a mess, especially this time of year. Sticks to everything like laser toner or those infernal foam packing peanuts.
(Score: 2) by krishnoid on Thursday January 31 2019, @08:14AM
Open-ended question -- how do QA groups at memory manufacturers identify the failure mode at the molecular/layer/cell level when it comes to flash memory?