For anyone interested in the ever expanding use cases for...

For anyone interested in the ever expanding use cases for weebits ReRAM a paper was recently submitted to arXiv (a non peer reviewed source) which concludes that their presented MemFHE FHE (Fully Homomorphic Computing) accelerator for client and server side FHE utilising PIM (Processing In Memory) with ReRAM which "reduces the data transfer bottlenecks... enables extensive parallelism... raises the bar of todays systems security, providing both classical and quantum-safe security guarantees." source: https://arxiv.org/abs/2204.12557

There are no current mass produced chips that can fill the RRAM role required by this state of the art encryption system. Weebits ReRAM seems like it would be perfect for this role, a form of encryption that will be integral in protecting data from quantum decryption in the future.

FHE is touted as the "holy grail" of security, see here: https://www.sciencedirect.com/topics/computer-science/fully-homomorphic-encryption

However FHE is not currently widely adopted due to it requiring 50-60x the computing power and 10-20x the memory compared to unencrypted soputions, which ReRAM should help to curb in an implementation like MemFHE. MemFHE, and implementations like it, should allow for significant speed increases over standard CPU (20Kx) and GPU (265x) implementations as well as increasing data throughput while implementing neural nets with FHE when compared to other state-of-the-art implementations of FHE.

I am by no means an expert in these fields, and if anyone with more security or big data related knowledge has anything to add or correct please do, I'm just excited to see another huge potential cloud storage and data centre related use case for Weebit's tech, provided we can actually produce an NVMe competitor.