Recent MAC improvements: Test Results

Chris Rapier rapier at
Tue Jun 12 08:46:49 EST 2007

I ran it a few times earlier with aes128 but the results were 
inconsistent. I felt it was because I was hitting a disk i/o bottleneck 
instead of a processor limitation. I'll try it now with arcfour256.

Still, even though the difference posted might not seem impressive I 
actually thought it was. 1MB/s can make a *serious* difference when you 
are dealing with terabyte datasets.

Damien Miller wrote:
> On Mon, 11 Jun 2007, Chris Rapier wrote:
>> Environment:
>> OpenSSH 4.6p1
>> 1Gb/s 1.12ms RTT
>> Linux 2.6.18-web100 to Linux 2.6.16-web100
>> Autotuning enabled. 2.4Ghz Xeon SMP
>> 20 iterations of 2GB scp transfer. Disk to /dev/null
>> scp -caes256-cbc -P 42222 ~/2gb rapier at delta:/dev/null
>> MAC - avg 25.0MB/s (1:22)
>> STD - avg 24.0MB/s (1:25)
> You will see the difference more clearly if you use a faster cipher,
> like aes128-cbc, aes128-ctr or arcfour256.
> -d

More information about the openssh-unix-dev mailing list