Samsung claims key-value Z-SSD will be fastest flash ever

Among a blizzard of news from Samsung's Tech Data, El Reg has spotted smaller processor nodes, FPGAs added to SSDs, stacked and cubed memory, quad-level cell flash and object-storing SSDs on the way. 7LPP Sammy has started fabbing 7LPP – 7nm extreme ultra-violet (EUV) low power plus lithography – wafers at its S3 Fab in …

[Flakk]

Sammy has started fabbing 7LPP – 7nm... and claims to now have a clear path of 3nm processes.

How do you say "pwned" in Korean? Asking for a friend... at Intel.

[Destroy All Monsters]

Frankly this is amazing. Will they REALLY reach 3nm?

[CheesyTheClown]

Yes please

Just... yes please.

I’ve been desperately waiting for a something like this. If they have a KV solution which supports replication that would be absolutely amazing!!!

[TheGreatCabbage]

It's getting quite exciting with upcoming 7nm Ryzen CPUs, 7nm Snapdragons and now Samsung's plans for 3nm processes!

[disk iops]

KV store

I'm guessing it's really just an optimization of what a traditional filesystem does. Instead of the OS requesting an arbitrary series of blocks based on it's own housekeeping records of 'path/file' which maps to a series of inodes and from there a series of block IDs, the OS just asks for a 'key' and the SSD has it's own list of extents/blockIDs that map to the 'value'. So instead of all that record-keeping at the OS level, all it needs to do is hash(path/file) and send that result to the storage device.

This removes at least one look-up table being maintained by the SSD and probably gives it flexibility in moving things around. It's probably very efficient on linear read/write but probably sucks on partial writes akin to the RAID-hole.

I could see AWS/S3 using these in the Data tier since Amazon already implements it like that - storage node manages it's in-chassis storage as key (blockID) -> list of device::extents[] that only it knows internally.