Almost on every post comparing DSA and RSA, I see a point brought up: DSA is faster for signature generation, but slower for validation. I wonder why this is?

For example, in DSA, to sign something, you have to do the same mod_pow as RSA, but then you also need to generate a hash of the message, and then calculate the inverse mod... With all those steps, why is DSA faster at signing than RSA?

  • (RSA also needs "to generate a hash of the message".) ​ ​ – user49075 Dec 9 '15 at 19:22
  • 1
    At current sizes (mostly 2k), DSA sign isn't very much faster than RSA sign: compare the size of k to the size of d, then consider that RSA privatekey can and usually does use the CRT optimization. But you may have misread; a point that is often made is that with RSA verify is much more than 10 times faster than sign, while with DSA they are nearly the same (verify is maybe 1.25 slower). This matters when the signer and verifier are different, such as the public web where each server may handle thousands to millions of clients. – dave_thompson_085 Dec 10 '15 at 10:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.