I think you aren't considering how much cpu load is added by opening a new circuit ... This latency you will experience is exactly the sort of thing that will get worse if people start overloading the network with extra circuits.

As with the current count of onions, it's not intentional harm, just people wanting to do things. Bitcoin is bringing a big growth in onion sites (litle impact itself). But the new bitcoin version can interop with clearnet and onions natively. If I remember from my last test, bitcoin rotates its traffic through a sizable address pool. Multiply that by the userbase and it could be similar to torrenting. There might be utility in allowing different circuit expiry parameter settings for different socksports... perhaps keeping a large state table for some would reduce load. Related at a distance is that I think the Phantom project mentioned they will not hesitate to use disk to support the network protocol.

While I'm at it, there *are* several steps that would lead to significantly improving hidden service performance: https://trac.torproject.org/projects/tor/ticket/1944 plus the various performance and security fixes in the 'Tor hidden service' category.

I want to get around to make some nice HS metrics along that ticket, but I don't want to give early hopes.

I know it can be tempting to treat the Tor design and code as a black box and try to hack around it, but I think in this case the clear right thing to do is to make the code not suck so much.

Yes this is nicer. _______________________________________________ tor-talk mailing list tor-talk@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk ----- End forwarded message ----- -- Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE