On Wed, Nov 19, 2014 at 11:59:50AM -0500, joe wrote: > But to get to your point on the performance "hit", when I benchmarked > AES-128 vs. AES-256 in OpenSSL with the openssl speed command, I just > didn't see a huge level of pain: > > The 'numbers' are in 1000s of bytes per second processed. > > type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes > aes-128 cbc 110889.82k 120032.96k 123677.01k 125399.38k 125501.99k > aes-256 cbc 81992.44k 87824.68k 88246.53k 88979.11k 89647.79k
Please note that you're not using the evp interface there, and so might for instance not be using AES-NI even your hardware has support for it. On my box without -evp: aes-128 cbc 110625.76k 119842.40k 123058.01k 123550.72k 124364.97k aes-192 cbc 94271.34k 101065.96k 102279.08k 102773.76k 103331.53k aes-256 cbc 79115.51k 86396.01k 87845.42k 88026.11k 88124.07k rc4 420860.77k 712188.45k 823340.46k 860657.32k 871604.14k camellia-128 cbc 101524.14k 154703.91k 175746.65k 181108.05k 182960.13k camellia-192 cbc 88263.40k 121438.20k 132575.74k 136623.10k 137639.25k camellia-256 cbc 88055.58k 121427.14k 132664.66k 136543.57k 138179.04k With -evp: aes-128-cbc 665401.82k 705108.14k 725408.17k 728927.91k 731252.44k aes-192-cbc 559985.05k 597358.49k 604821.50k 601399.31k 608561.83k aes-256-cbc 482553.42k 513640.24k 520753.75k 521101.31k 524263.34k rc4 387369.55k 677893.89k 816945.83k 856821.76k 869501.61k camellia-128-cbc 96466.97k 151111.64k 174039.13k 178602.33k 182517.21k camellia-192-cbc 83364.92k 119781.63k 132168.28k 136482.47k 137655.64k camellia-256-cbc 83568.77k 119611.78k 132230.49k 136427.18k 137620.14k Kurt _______________________________________________ Uta mailing list [email protected] https://www.ietf.org/mailman/listinfo/uta
