Delivery-Date: Thu, 25 Feb 2016 19:11:49 -0500
Return-Path: <tor-talk-bounces@lists.torproject.org>
X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on moria.seul.org
X-Spam-Level: 
X-Spam-Status: No, score=-4.2 required=5.0 tests=BAYES_00,RCVD_IN_DNSWL_MED,
	T_RP_MATCHES_RCVD autolearn=ham version=3.3.1
X-Original-To: archiver@seul.org
Delivered-To: archiver@seul.org
Received: from eugeni.torproject.org (eugeni.torproject.org [38.229.72.13])
	(using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits))
	(No client certificate requested)
	by khazad-dum.seul.org (Postfix) with ESMTPS id CCAA51E04BA;
	Thu, 25 Feb 2016 19:11:47 -0500 (EST)
Received: from eugeni.torproject.org (localhost [127.0.0.1])
	by eugeni.torproject.org (Postfix) with ESMTP id 7401B39693;
	Fri, 26 Feb 2016 00:11:41 +0000 (UTC)
Received: from localhost (localhost [127.0.0.1])
 by eugeni.torproject.org (Postfix) with ESMTP id 086343968F
 for <tor-talk@lists.torproject.org>; Fri, 26 Feb 2016 00:11:37 +0000 (UTC)
X-Virus-Scanned: Debian amavisd-new at 
Received: from eugeni.torproject.org ([127.0.0.1])
 by localhost (eugeni.torproject.org [127.0.0.1]) (amavisd-new, port 10024)
 with ESMTP id 8Wdzp_7umnPo for <tor-talk@lists.torproject.org>;
 Fri, 26 Feb 2016 00:11:36 +0000 (UTC)
Received: from meiko.romanrm.net (meiko.romanrm.net
 [IPv6:2001:bc8:3829:100::1])
 (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
 (Client did not present a certificate)
 by eugeni.torproject.org (Postfix) with ESMTPS id CE7D9393E9
 for <tor-talk@lists.torproject.org>; Fri, 26 Feb 2016 00:11:36 +0000 (UTC)
Received: from natsu (unknown [IPv6:fd39::e9:9eff:fe8f:1bcf])
 by meiko.romanrm.net (Postfix) with SMTP id CBC2E11AF2C;
 Fri, 26 Feb 2016 00:11:04 +0000 (UTC)
Date: Fri, 26 Feb 2016 05:11:04 +0500
From: Roman Mamedov <rm@romanrm.net>
To: Ryan Carboni <ryacko@gmail.com>
Message-ID: <20160226051104.5a244201@natsu>
In-Reply-To: <CAO7N=i2t4xF7K46xi==XM=C2Nk-HhJj4Ywtp+0J8ee_b7xbDEQ@mail.gmail.com>
References: <CAO7N=i2t4xF7K46xi==XM=C2Nk-HhJj4Ywtp+0J8ee_b7xbDEQ@mail.gmail.com>
X-Mailer: Claws Mail 3.11.1 (GTK+ 2.24.10; x86_64-pc-linux-gnu)
MIME-Version: 1.0
Cc: tor-talk@lists.torproject.org
Subject: Re: [tor-talk] FPGA Tor Relay
X-BeenThere: tor-talk@lists.torproject.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: tor-talk@lists.torproject.org
List-Id: "all discussion about theory, design,
 and development of Onion Routing" <tor-talk.lists.torproject.org>
List-Unsubscribe: <https://lists.torproject.org/cgi-bin/mailman/options/tor-talk>, 
 <mailto:tor-talk-request@lists.torproject.org?subject=unsubscribe>
List-Archive: <http://lists.torproject.org/pipermail/tor-talk/>
List-Post: <mailto:tor-talk@lists.torproject.org>
List-Help: <mailto:tor-talk-request@lists.torproject.org?subject=help>
List-Subscribe: <https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk>, 
 <mailto:tor-talk-request@lists.torproject.org?subject=subscribe>
Content-Type: multipart/mixed; boundary="===============0210473875069622302=="
Errors-To: tor-talk-bounces@lists.torproject.org
Sender: "tor-talk" <tor-talk-bounces@lists.torproject.org>

--===============0210473875069622302==
Content-Type: multipart/signed; micalg=pgp-sha1;
 boundary="Sig_/ctRN0pOizEfhwrq20PUP/p8"; protocol="application/pgp-signature"

--Sig_/ctRN0pOizEfhwrq20PUP/p8
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

On Thu, 25 Feb 2016 12:21:42 -0800
Ryan Carboni <ryacko@gmail.com> wrote:

> http://netfpga.org/site/#/systems/1netfpga-sume/details/
>=20
> This is apparently available for an academic price of around two
> thousand dollars.

And then you have a programming job on your hands to adapt Tor into the FPG=
A,
that could cost one or two magnitudes more than that...

> Such cards will probably have to be used in the near future, at least
> to reduce bandwidth costs per gigabit/s speed.

Maybe I'm missing something, how anything you do inside your server (run Tor
on CPU, GPU, FPGA or magic fairies) will reduce your *bandwidth* costs?

> I imagine infrequent operations and operations likely to be changed
> between Tor versions could be offloaded to the CPU.

- Tor is already sped up immensely if you use a CPU which has the hardware =
AES
  acceleration, i.e. almost any modern x86 CPU. (Not sure if there are any
  other operations you could offload to FPGA, or if FPGA could be faster th=
an
  an AES-NI CPU at AES.)

- ...then you could optimize Tor to use more than ~1.3-1.5 of a CPU core at
  most as it does currently to scale further, as many modern CPUs easily ha=
ve
  6-8 cores. (This is likely easier than rewriting it to use FPGA). As a
  stop-gap measure, people with such CPUs currently have to run two instanc=
es
  of Tor per each IPv4 address their server has.

In the end, if you could just fully load 8 cores of a humble $170 AES-NI CP=
U, I
believe this should be already enough to process a full gigabit of traffic,=
 or
even more. And you don't really need much more, since it is rare that you c=
an
reasonably get more than a gigabit of unmetered bandwidth per single server.

--=20
With respect,
Roman

--Sig_/ctRN0pOizEfhwrq20PUP/p8
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iEYEARECAAYFAlbPmBgACgkQTLKSvz+PZwhIKQCfXWFqoX1UnOBJOyEw17OaOvfY
iGgAnRavldRYFccdKic+ydgfkY3jDRJT
=azou
-----END PGP SIGNATURE-----

--Sig_/ctRN0pOizEfhwrq20PUP/p8--

--===============0210473875069622302==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

-- 
tor-talk mailing list - tor-talk@lists.torproject.org
To unsubscribe or change other settings go to
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk

--===============0210473875069622302==--

