Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email firstname.lastname@example.org
From: John P. Sullivan (jsullivanUS.NCIPHER.COM)
Date: Mon Apr 30 2001 - 15:32:04 CDT
You might also want to look into using nCipher HSMs:
I can get you doc (or answer your questions) on our APIs if you'd like...
> -----Original Message-----
> From: Secure Programming Mailing List
> [mailto:SECPROGSECURITYFOCUS.COM]On Behalf Of Seth Arnold
> Sent: Monday, April 30, 2001 4:00 PM
> To: SECPROGSECURITYFOCUS.COM
> Subject: Re: RSA
> [Adam, please wrap your lines at 72 characters. Thanks.]
> * Adam Berent <adminIDEVELOPERONLINE.COM> [010430 09:15]:
> > When implementing the RSA algorithm how do you handle large numbers
> > such as a 448 bit modulus. Aren't the unsigned long data types too
> > small for that. Is there a special number math library that people
> > use?
> Unsigned long is often 64 bits. One comes up very short with this few
> bits. GMP is one answer. However, if you are thinking of implementing
> RSA yourself, I would ask you to consider using OpenSSL's RSA support
> before you start work. Chances are good they will have done a better job
> than you in their implementation. Another option is Wei Dai's Crypto++.
> His license is often very amenable to companies. :)
> If speed is an issue, look into IBM's 4758 cryptography accelerator. Sun
> has a similar product. Both companies should provide APIs for their
> products that you can use. They likely use their own arbitrary precision
> libraries to take advantage of the hardware.
> I hope this helps.
> Earthlink: The #1 provider of unsolicited bulk email to the Internet.