[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: performance tuning...
On Mon, Apr 11, 2005 at 10:45:47AM -0400, Jonathan D. Proulx wrote:
> I'm wondering what if anything I can due to tune my system (spoon) for
> maximum tor performance (and I suppose how to judge what that is).
> This may be a bit gratuitous since it's usually in the top 3 as listed
> on http://serifos.eecs.harvard.edu:8000/cgi-bin/exit.pl?sortbw, but
> I'm confused by it's wandering bandwidth claims, from about 1M -> 1.5M
> as this doesn't seem to relate to system load.
From the tor-spec.txt (documenting the descriptor line):
"bandwidth" bandwidth-avg bandwidth-burst bandwidth-observed
Estimated bandwidth for this router, in bytes per second. The
"average" bandwidth is the volume per second that the OR is willing
to sustain over long periods; the "burst" bandwidth is the volume
that the OR is willing to sustain in very short intervals. The
"observed" value is an estimate of the capacity this server can
handle. The server remembers the max bandwidth sustained output
over any ten second period in the past day, and another sustained
input. The "observed" value is the lesser of these two numbers.
for a bit more info.
> I'm running 0.1.0.3-rc on Linux (Debian Sid) on a fast machine (dual
> 2.4GHz Xeon 2G RAM) with a fast network connection (1G localally with
> dual 100M exits to the internet at large).
> Tor currently uses a good bit of CPU, but not to the point it
> interferes with the normal operation of this machine, I'd liek to get
> it to the point that it does and then back off from there a bit.
> So any dark voodoo in syctl settings or anything that people have to
> offer? Would it be worth recompiling with -march=pentium4 or other
Let us know what you find. The 0.1.0.3-rc release has some more
cpu improvements, but the main pain comes from crypto, and that is
proportional to the number of clients who are asking you to handle a
circuit for them. In any case, you won't be filling up all 100Mb yet --
see the lower graph of http://new.noreply.org/tor-running-routers/
Also check out