From owner-freebsd-net@freebsd.org Sat May 20 00:56:29 2017 Return-Path: Delivered-To: freebsd-net@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 9BD06D73634 for ; Sat, 20 May 2017 00:56:29 +0000 (UTC) (envelope-from jordancaraballo87@gmail.com) Received: from mail-ua0-x22c.google.com (mail-ua0-x22c.google.com [IPv6:2607:f8b0:400c:c08::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 50C3B171D for ; Sat, 20 May 2017 00:56:29 +0000 (UTC) (envelope-from jordancaraballo87@gmail.com) Received: by mail-ua0-x22c.google.com with SMTP id u10so34877168uaf.1 for ; Fri, 19 May 2017 17:56:29 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:subject:to:message-id:date:user-agent:mime-version :content-transfer-encoding; bh=/AXKisq+0/Ab2rkkVnPIRVFuJjkN6OIhLLZdbHqodlk=; b=mlCl44bLWB5IHPZXE/KjA/64vY/ihCMWl6wsivHR+jnr/+XqtaOB2rlFDtYlwQQn5I taRPBIVo3OL76GxO0lnpf8dr/gHwDEFoXb+Bgi78DyD97uPtrdltPvkHJdpCibX9FZlE S9n7vDWxxfix5mQ1tjSib7qNdL1XR/X7WbWNErecz7zQZWK7uKXrt5g3aJeKXLiBa3aW fHDjj86Q4R/qMNCWyusArHaQadKDVYGuy5PxFMyy7Yon9BqoECSF4e6FB/EYjz20C/xv /wA9d9Y98VHt2VjMiyG4cHWXh5B7Y+lzEiPNNJ9u6P757L4fMC7zz7WQIbzfCutMB/1C zqKw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:subject:to:message-id:date:user-agent :mime-version:content-transfer-encoding; bh=/AXKisq+0/Ab2rkkVnPIRVFuJjkN6OIhLLZdbHqodlk=; b=B8QTk0E3PGKt4OqrDBpXSWPgMYcJO1aDqI6+j3RPwzjx1vQAnvxnzzpvejha24QNR9 tEsmYALmNSM8aaxmdV8YHl3TQrNUqpGSMo/reJriYpwx9PipApZVmPFKKDSEZn6/Cdd5 1is/LHhT/JkL/DfLCHW/b5Idy6NiUlHnN0mMV1JWQtEfIYLeQACpxwurCwyzM/ui70No ik/z3zHKyPUat+KlpXYiCltNcBbg2826D1mlP+woRp2ASzQJLIZuAsGi2S8YlviVpCOb V0w7Af1KjQNekl9dLSNBYuYkfbKNR+vNHuYdXFGA5KRQnc2uHdjbyG1cqUPmx4RAY+Io kimg== X-Gm-Message-State: AODbwcBKJ+1WF/rCT1sztNJE3E7tLttvbWNnqroT4lwqcfwXsB7+h78Q 4EHGDdZZXy5KUg== X-Received: by 10.176.87.217 with SMTP id u25mr6209247uac.90.1495241788117; Fri, 19 May 2017 17:56:28 -0700 (PDT) Received: from [10.0.0.66] ([173.215.201.67]) by smtp.gmail.com with ESMTPSA id 8sm1177217vki.32.2017.05.19.17.56.26 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 19 May 2017 17:56:27 -0700 (PDT) From: "Caraballo-vega, Jordan A. (GSFC-6062)[COMPUTER SCIENCE CORP]" Subject: What is the status of the freebsd routing proposal within -stable and -current? Dell PE R530 passing 8.4M pps To: freebsd-net@freebsd.org Message-ID: Date: Fri, 19 May 2017 20:56:34 -0400 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.11; rv:45.0) Gecko/20100101 Thunderbird/45.7.0 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 May 2017 00:56:29 -0000 Hi, Below are some "netstat -w1 -d -h" results of a Dell PE R530 system (2 Intel(R) Xeon(R) E5=C2=AD2695 CPU's, 18 cores per cpu), equipped with a Chelsio T-580-CR dual port in an 8x slot. Versions 12-CURRENT (nodebug), 11-STABLE, and FreeBSD Routing Proposal image (http://dev.bsdrp.net/benchs/nanobsd/fbsd11-routing.r287531/) were used. This image was used to reproduce these results http://blog.cochard.me/2015/09/receipt-for-building-10mpps-freebsd.html. As a summary: -CURRENT(nodebug) ~5.0M pps, -STABLE ~5.8M pps, and RP ~8.4M pps. Results are shown below. Additional configurations and stats can be given upon request. 1. What is the merge/implementation status of this routing proposal proje= ct? 2. Suggestions and feedback to push more pps and improve these scores are more than welcome. Testing environment: * Switch with 48x 10G ports and 12x 40G ports was used=E2=80=A8, (48) 10G= connected nodes were used, (24) nodes on each side of the firewall=E2=80=A8= , Small packet per second (PPS) tests were run using 'iperf' (64 bytes UDP)= =2E Following indications from Navdeep Parhar that UDP queue hashing is not as efficient as it could be, we started running the tests with various powers of 2 streams (2,4,8,16,32,100 per system). -CURRENT (nodebug) input (Total) output packets errs idrops bytes packets errs bytes colls drop= s 4.8M 0 0 503M 4.8M 0 503M 0 0= 5.1M 0 0 536M 5.1M 0 536M 0 0= 4.9M 0 0 511M 4.9M 0 511M 0 0= 4.9M 0 0 510M 4.9M 0 510M 0 0= 4.7M 0 0 495M 4.7M 0 495M 0 0= 5.0M 0 0 528M 5.0M 0 528M 0 0= -STABLE 5.1M 2.9k 0 536M 5.1M 0 536M 0 0 5.8M 3.2k 0 607M 5.8M 0 607M 0 0 4.2M 2.4k 0 444M 4.2M 0 444M 0 0 5.8M 3.3k 0 607M 5.8M 0 607M 0 0 5.8M 3.3k 0 607M 5.8M 0 607M 0 0 5.8M 3.2k 0 606M 5.8M 0 606M 0 0 5.6M 3.2k 0 591M 5.6M 0 591M 0 0 -Routing Proposal 12M 0 2.9M 1.2G 8.4M 0 881M 0 465k= 12M 0 2.9M 1.2G 8.4M 0 880M 0 457k= 12M 0 2.9M 1.2G 8.4M 0 879M 0 488k= 12M 0 2.9M 1.2G 8.4M 0 881M 0 476k= 12M 0 2.9M 1.2G 8.4M 0 879M 0 473k= 12M 0 2.9M 1.2G 8.4M 0 883M 0 482k= Thanks in advance, - Jordan