4 ms·
I wonder if doing something similar to Ciphercloud, using a homomorphic encryption library like libScarab[0], would actually make it secure. I guess I still don
by jared314 14y ago
I wonder if doing something similar to Ciphercloud, using a homomorphic encryption library like libScarab[0], would actually make it secure. I guess I still don't understand what Ciphercloud does.
[0] https://hcrypt.com/scarab-library/ https://hcrypt.com/scarab-library/
- betterunix 14y agoFHE is far too slow for any practical use. In a few more years, things might be different, but for now anyone marketing a practical FHE solution is probably lying.
- jared314 14y agoI'm still trying to find performance numbers that prove it impractical. Otherwise, it just sounds like a problem that could be mitigated by clustering.
- betterunix 14y agoHere are some results from a research team that has been on the forefront of FHE implementations; note that this has been improved on significantly since last August, but you are still looking at minutes of computation for relatively small functions: http://www.iacr.org/cryptodb/archive/2012/CRYPTO/presentation/17-1-Smart.pdf http://www.iacr.org/cryptodb/archive/2012/CRYPTO/presentatio... Also, throwing "clustering" at every problem is misguided. Not all problems are easily parallelized: https://en.wikipedia.org/wiki/P-complete https://en.wikipedia.org/wiki/P-complete
- jared314 14y agoDon't forget Amdahl's law and Gustafson's law on the limits of parallelizing when the problem isn't P-complete. Either way, I disagree with your conclusion that it is misguided.