From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-vc0-x22c.google.com (mail-vc0-x22c.google.com [IPv6:2607:f8b0:400c:c03::22c]) by dpdk.org (Postfix) with ESMTP id CA3269DE for ; Wed, 4 Dec 2013 22:48:09 +0100 (CET) Received: by mail-vc0-f172.google.com with SMTP id hz11so12206408vcb.3 for ; Wed, 04 Dec 2013 13:49:12 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=TgBOE5vJglExdQLWDl3DXxGJaXvkgPGwcdy6fOynyYM=; b=jvnrzQ5eLAHO2A1EuYHXEYu912fxd6NwMlpMPBjSl/tcbX7EFjLrclk1fPoHifUiwl uB2Q3DCHz+CWtfIxkqMJ8HtiuPo3DQ7y9KGG/uojN8rBP2Se/P2JM58NUcj6stLpEdmZ BKbPi/7fbvQ7KTwcWf2LfHu0lD238B7in3DjDt64JCzeLRbVd26T4ObtE4bIn7yWiZRB 5XjIUeIH9XYb6VdUZLq4YOXEIdY2/5GcxLsgNM2610aoWOI4C/QwuAx5gJnzkfQ+QNgc 0slBe7z3IjKigtnlMow4OHB72OQEnWrpttqXJcKuGSXLKOMGQN8SRt2rTEl4hvRH6Uhw sNCw== MIME-Version: 1.0 X-Received: by 10.52.164.16 with SMTP id ym16mr2032512vdb.39.1386193752158; Wed, 04 Dec 2013 13:49:12 -0800 (PST) Received: by 10.52.177.70 with HTTP; Wed, 4 Dec 2013 13:49:12 -0800 (PST) In-Reply-To: <03ee01cef12b$c56d85a0$504890e0$@com> References: <03b201cef0f7$5fba0d20$1f2e2760$@com> <03ee01cef12b$c56d85a0$504890e0$@com> Date: Wed, 4 Dec 2013 13:49:12 -0800 Message-ID: From: Sambath Kumar Balasubramanian To: =?ISO-8859-1?Q?Fran=E7ois=2DFr=E9d=E9ric_Ozog?= Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.15 Cc: "dev@dpdk.org" Subject: Re: [dpdk-dev] Question on the Ring Library X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 04 Dec 2013 21:48:10 -0000 Thanks Fran=E7ois-Fr=E9d=E9ric. Trying to embark on a small prototype and s= ee the results. Thanks for the timing data. Really helpful. Regards, Sambath On Wed, Dec 4, 2013 at 12:02 PM, Fran=E7ois-Fr=E9d=E9ric Ozog = wrote: > You are welcome ! > > > > Even if you insert packets in batch into a fifo, the mutex is still > unpredictable. If one pthread_lock costs 1ms, you are going to lose > packets, regardless of the number of RSS queues and ring sizes=85 Batchin= g > comes with another issue: need to flush a batch after a certain timeout i= f > you do not have packets. Userland timer resolution is such that you are > introducing port to port latency quite a lot: DPDK native, you can manage > to get a 2=B5s latency (including PCI express), while with batch it may b= e > 1ms=85 You may find smarter algorithms but still you will be an order of > magnitude higher than DPDK. > > > > Hope this helps. > > > > Fran=E7ois-Fr=E9d=E9ric > > > > *De :* Sambath Kumar Balasubramanian [mailto: > sambath.balasubramanian@gmail.com] > *Envoy=E9 :* mercredi 4 d=E9cembre 2013 15:01 > *=C0 :* Fran=E7ois-Fr=E9d=E9ric Ozog > *Cc :* dev@dpdk.org > *Objet :* Re: [dpdk-dev] Question on the Ring Library > > > > Thanks Fran=E7ois-Fr=E9d=E9ric. That puts real good perspective on the co= st for > the vent assuming each packet in the fast will result > > in an event. If event rate is orders of magnitude less than the packet > rate, then I guess we can still achieve 10G since the "extra cost" will b= e > in the event thread and not in the fast path thread. > > > > Regards, > > Sambath >