2006 | OriginalPaper | Buchkapitel
A Fast RSA Implementation on Itanium 2 Processor
verfasst von : Kazuyoshi Furukawa, Masahiko Takenaka, Kouichi Itoh
Erschienen in: Information and Communications Security
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
We show the fastest implementation result of RSA on Itanium 2. For realizing the fast implementation, we improved the implementation algorithm of Montgomery multiplication proposed by Itoh et al. By using our implementation algorithm, pilepine delay is decreased than previous one on Itanium 2. And we implemented this algorithm with highly optimized for parallel processing. Our code can execute 4 instructions per cycle (At maximum, 6 instructions are executed per cycle on Itanium 2), and its probability of pipeline stalling is just only 5%. Our RSA implementation using this code performs 32 times per second of 4096-bit RSA decryption with CRT on Itanium 2 at 900MHz. As a result, our implementation of RSA is the fastest on Itanium2. This is 3.1 times faster than IPP, a software library developed by Intel, in the best case.