aes performance
Scott Bennett
bennett at cs.niu.edu
Wed Feb 25 22:19:27 UTC 2009
On Wed, 25 Feb 2009 22:59:14 +0100 Olaf Selke <olaf.selke at blutmagie.de>
wrote:
>I wrote:
>>
>> as I understood tor spends most of its cpu time within openssl library aes crypto.
>> Which result of "openssl speed aes" applies to tor? Is it aes-128 cbc 16 bytes?
>> In this case my old Prestonia P4 Netburst Xeon box's throughput is supposed to
>> be roughly about 40 MBit/s as middleman. Correct?
>>
>> type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
>> aes-128 cbc 84098.99k 119729.69k 138053.97k 142741.16k 144386.04k
>> aes-192 cbc 75035.35k 104143.72k 115681.81k 120099.84k 120949.42k
>> aes-256 cbc 69559.47k 92221.78k 102006.05k 105361.75k 100274.74k
>
>stupid! I mixed up openssl's benchmark bytes/s from the table above with
>bits/s network throughput. Now it appears only 10% tor's cpu usage is
>spent within aes crypto. What the heck is tor doing the remaining 90%?
>
Olaf, your system is one of 14 or 15 tor relays that routinely peak at
more than 5 MB/s. You have, in the past, stated that it typically has ~5,000
established connections simultaneously. I would guess that key exchange
operations on a system loaded like that must take up an awful lot of CPU time.
My relay peaks at about 1/10 of blutmagie's peaks. tor's primary CPU
percentage range seems to be 0% - 13% (2-second averages) on my 3.4 GHz P4
Prescott (which is also a Netburst chip). Occasionally, it will consume more
than that for a brief time, sometimes as much as 30% - 40%, but those events
usually last only a couple of seconds. Given these numbers, your figures seem
reasonable.
Scott Bennett, Comm. ASMELG, CFIAG
**********************************************************************
* Internet: bennett at cs.niu.edu *
*--------------------------------------------------------------------*
* "A well regulated and disciplined militia, is at all times a good *
* objection to the introduction of that bane of all free governments *
* -- a standing army." *
* -- Gov. John Hancock, New York Journal, 28 January 1790 *
**********************************************************************
More information about the tor-talk
mailing list