• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Future of flash

Brian Stirling

Diamond Member
With each new generation of flash the features sizes get smaller permitting greater density and storage, but it also comes with the penalty of shorter life owing to the limited number of write cycles. Going forward it would seem the lifespan will quite soon get to be an actual problem that limits further density increases.

One technique I've heard about that appears to undue the damage is thermally cycling the device but if that works, how could this be implemented to repair a device in situ? If the thermal cycle can be of short duration it's possible the modules/chips could have a heater built into them and, periodically, the devices are heated, but, I would imagine that you'd have to assume all memory to be lost in the process.

So, has anyone heard more on this...


Brian
 
I don't have specifics, but maybe they have some way to provide a heater to each chip, and then just apply heat to one chip at a time?

So maybe when the drive does its TRIM thing, it can shuffle all data off one entire chip, heat cycle it to make it fresh/new, then TRIM the data back to it, and go through all chips eventually? I think separate heaters can be very small, maybe like a sheet that is adhered to a chip.

Another option would be to just use the chip itself to generate heat at each cell junction or something, so you only affect the desired cell and maybe a few neighbors where, again, you can shuffle data around on-the-fly to avoid losing anything you want, and cascade the heating location around as you move data out of the way dynamically.
 
Back
Top