I am currently peaking 46.39 MBps / 371.12Mbps, https://atlas.torproject.org/#details/F14B7BF44F9B170DFF628F237E0C7E8D631F957E, with "NumCPUs 2" on an AMD A8-5600K with 8GB RAM. My setup is based on https://wiki.archlinux.org/index.php/Tor#.2B100Mbps_Exit_Relay_configuration_example.

I presume you have read https://blog.torproject.org/blog/lifecycle-of-a-new-relay, so you know it will take some time before the network takes full advantage of your new exit relay.

On 2 January 2015 at 17:03, Austin Bentley <ab6d9@mst.edu> wrote:

Actually, on 2nd thought, you may not have to limit your bandwidth because Tor MAY handle this for you.

Also, it's recommended to run your (presumably, 8) servers on different network addresses as well. If you are running a colocation rack this won't be difficult, but if you are doing this from your home.. well.. I won't start.

On Fri, Jan 2, 2015 at 10:00 AM, Austin Bentley <ab6d9@mst.edu> wrote:
Yep, you got it. Multiple processes with different configurations. You should also limit their bandwidths proportionally so you don't saturate your network interface.

On Fri, Jan 2, 2015 at 9:48 AM, Kura <kura@kura.io> wrote:
Hey guys, I recently decided to get myself an 8 core, 16 GB RAM machine to use for running an exit relay and was wondering, Tor only works on one core, even setting NumCPUs to 2 doesn't do a whole lot so, how is it even possible to get more than maybe, 300Mbps or so from one relay? Maybe I'm missing something but, running multiple Tor processes is just going to have multiple relays with different OR and Dir ports for each, right?

--
Kura

t: @kuramanga
w: https://kura.io/
g: @kura

_______________________________________________
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays

_______________________________________________
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays