RSA 1024 ridiculous

Sven Radde sven at
Sun Jun 17 20:38:52 CEST 2007


Andrew Berg schrieb:
> Try signing/encrypting files that are tens, hundreds, or thousands of
> megabytes in size. Sure, your average machine can sign/encrypt
> messages that don't even fill a cluster without breaking a sweat, but
> if the sensitive data is large, RSA-4096 isn't a good choice unless a
> gov't agency wants that data.

No matter what size the data is that you want to encrypt/sign, the size
of the public key only adds a constant factor to it.

The actual "bulk" data processing is done by a symmetric algorithm /
hash function. You only encrypt the key to the symmetric algorithm /
sign the hash value. Both are typically 256bit or smaller.

In fact, the larger the data you want to process, the *smaller* the
impact of a larger key is. (If it takes minutes to hash a few gigabytes,
it doesn't matter if signing the hash takes 10, 100 or 1000 milliseconds.)
Using email actually is something like the "worst case" for large public

You may want to do some research on "hybrid" cryptosystems for more
thorough information.

cu, Sven

More information about the Gnupg-users mailing list