Print

Welcome to Smokey's Security Forums.

Guests have only limited access to the board and it's features, please consider registering to gain full access!

Registration is free and it only takes a few moments to complete.

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jetico]

Thank you!

"AESNI" is not a modification of  AES encryption algorithm or some
specific mode used with AES algorithm. It is a set of 6 new instructions
for Intel Westmere-32nm processor architecture ("AESNI" means
"Advanced Encryption Standard New Instructions").

At the moment we have not tested real performance of software that
use hardware acceleration with AESNI instructions. According to
estimations made by Intel, performance of such a software with
AESNI hardware acceleration may increase up to 1.5 - 2 times.

Please note also that overall performance of encryption software
depends also on performance of disk subsystem. Besides, the question
of encryption algorithm modes still exists, because CBC, LRW, XTS
(or other) work with significantly different speeds.

From the other hand, encryption software may utilize several algorithms.
For example, BestCrypt Volume Encryption uses also RC6, which as minimum
1.5 times faster than AES. So although it is nice that AES hardware
acceleration is coming, we should think carefully before saying that
revolutionary step forward will happen soon in performance of software
encryption products.

[crosscheck]

I have been reading about the supposed AES hardware acceleration on the new Intel mobile Arrandale processors. The articles say that software that is written to use the "AES-NI" algorithm on the processor will be able to encrypt and decrypt much faster than current software-only encryption.

Is Jetico currently planning to incorporate this hardware acceleration into the BestCrypt software? Has Jetico done any testing to see if there are any real speed benefits and what they might be?

Thanks and Happy New Year everyone!