Re:NSA announces plans for transitioning to quantum resistan...
by jackgavigan on Saturday August 15, 2015 @10:04AM
(#10065388)
> You'll note they don't actually specify any asymmetric quantum-resistant algorithms. I'd guess if they did, NTRU or a derivative would be one they'd consider first...
My bet would be on McEliece. It's been around longer, so has been subjected to more rigorous cryptanalysis than NTRU, and is not patented.
Re:NSA announces plans for transitioning to quantum resistan...
by eeZi on Saturday August 15, 2015 @08:33AM
(#10065217)
IPSec still has forward secrecy even with pre-shared keys.
Re:NSA announces plans for transitioning to quantum resistan...
by AnthonyMouse on Saturday August 15, 2015 @09:55AM
(#10065368)
The way it does that is by using Diffie-Hellman or ECDH, which both rely on the hardness of the discrete logarithm problem that quantum computers would break.
Re:NSA announces plans for transitioning to quantum resistan...
by wolf550e on Saturday August 15, 2015 @09:14AM
(#10065287)
openssl 1.0.2a on an old Conroe Core2 gets this:
sign verify sign/s verify/s
256 bit ecdsa (nistp256) 0.0001s 0.0003s 8727.8 3493.3
384 bit ecdsa (nistp384) 0.0005s 0.0020s 2001.4 493.0
521 bit ecdsa (nistp521) 0.0010s 0.0017s 1021.3 603.0
Can P-384 (or Ed448-Goldilocks) be close to P-256 in speed?
Re:NSA announces plans for transitioning to quantum resistan...
by tveita on Saturday August 15, 2015 @10:38AM
(#10065474)
Here are some median times from the eBATS benchmark page for a 2013 Intel Xeon E3-1275 V3 3500MHz (http://bench.cr.yp.to/web-impl/amd64-titan0-crypto_sign.html)
Cycles to generate a key pair:
ed448goldilocks: 176924
ecdonaldp256: 290628
ecdonaldp384: 2202380
Cycles to sign 59 bytes:
ed448goldilocks: 185056
ecdonaldp256: 381696
ecdonaldp384: 2367856
Cycles to verify 59 bytes:
ed448goldilocks: 583900
ecdonaldp256: 913848
ecdonaldp384: 2741028
(ecdonaldp is ECDSA signatures with NIST P-256/384. The implementation used is OpenSSL, though I don't know which version)AFAIK no elliptic curve size is quantum secure, so I guess the goal is just to require slightly more qubits for an attack.
Pre-shared keys, as they suggest there, have no forward secrecy - which makes them great for those who really like stealing, say, IPsec keys… like the NSA. It may work with the kind of old military key infrastructure they and GCHQ have, that regularly distributes random keys from centralised, organisationally-trusted sources on specialised hardware; it is a terrible recommendation for civilians.
Interesting that they're still married to P-384 (probably the most annoying curve to implement correctly). Properly-implemented Ed448-Goldilocks is safer, and that's what CFRG are going with for the "paranoid" level.