From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?iso-8859-1?Q?Fran=E7ois-Fr=E9d=E9ric_Ozog?= Subject: Re: Question on the Ring Library Date: Wed, 4 Dec 2013 21:02:32 +0100 Message-ID: <03ee01cef12b$c56d85a0$504890e0$@com> References: <03b201cef0f7$5fba0d20$1f2e2760$@com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Cc: dev-VfR2kkLFssw@public.gmane.org To: "'Sambath Kumar Balasubramanian'" Return-path: In-Reply-To: Content-Language: fr List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces-VfR2kkLFssw@public.gmane.org Sender: "dev" You are welcome ! =20 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 Batching comes = with another issue: need to flush a batch after a certain timeout if 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 be 1ms=85 You = may find smarter algorithms but still you will be an order of magnitude = higher than DPDK. =20 Hope this helps. =20 Fran=E7ois-Fr=E9d=E9ric =20 De : Sambath Kumar Balasubramanian [mailto:sambath.balasubramanian-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org]=20 Envoy=E9 : mercredi 4 d=E9cembre 2013 15:01 =C0 : Fran=E7ois-Fr=E9d=E9ric Ozog Cc : dev-VfR2kkLFssw@public.gmane.org Objet : Re: [dpdk-dev] Question on the Ring Library =20 Thanks Fran=E7ois-Fr=E9d=E9ric. That puts real good perspective on the = cost 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 be in = the event thread and not in the fast path thread. =20 Regards, Sambath