From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-x234.google.com (mail-qk0-x234.google.com [IPv6:2607:f8b0:400d:c09::234]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 848283BA8E for ; Sat, 28 Jul 2018 12:36:27 -0400 (EDT) Received: by mail-qk0-x234.google.com with SMTP id b5-v6so5301371qkg.6 for ; Sat, 28 Jul 2018 09:36:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc:content-transfer-encoding; bh=CSK92/ckMv9bimIDOS/qpbYEv1CY/A++p3uOx8PUjMw=; b=Vuy5Gu2lXDG6ljwg0KR+1C6GuS/uHRRprHWVEVDjj6Bw5EqccH0YzPA7t7+0Yv9Xjy iwon1dBQYI8MCjMJyAAhs1t6s2KCrk6PRHHKsyqfQRyXwZPmpTI/MH2ZDkcRvx94Fyao 2BwBKYB8fIeDDHO0ZL0PxxXRRLrKsX1TLXBDGsLaaKCtyOioTGBh87E8HPWpBq47Y445 6pFHVqLIgA5CrpU9IxssAEpyYVCJorfE+x9EuINfviXi0pxkVmfZEsk6QbdCN2tll/eR hb2iCP64tH50rIghCRpqdD9Etq4tCz2IWZE8Hwm9v/ILXDzsI/vxhwAWskmO16myqSDw ybtw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc:content-transfer-encoding; bh=CSK92/ckMv9bimIDOS/qpbYEv1CY/A++p3uOx8PUjMw=; b=RjQKsxfqR51P1XF+aTi0qUGQSj+lCJVfSLoWnvLFU7n0CTMQSXg6dUdLSpu3ghMhQ0 Y+ic5wvjSFINFzc2OtEBdEocolJ9MfIpj2rUbuFwaOu/aUhirxXFz+8eAb6yHY1C68o9 +mqKwITQ7hydh6rxdhXUb1qZLjYDg2PaS7Lw/Uel7+FFXA762pgAkTQ2c8ZfcL+lkIQE hSfzNjKo+Tb0ds/3yb66MNLYbiaSa2rhhNxhWFSE/CX9lhsj5HJym06+mr0OZw2nyDWd A5j8Aez/3LdibyLgtfcZnHjydvfrRMM8uBE7DiaXUS4ve8NUp459pZE9l5K3NOvSbzF9 J0hg== X-Gm-Message-State: AOUpUlHlsarBLkCRIMeMUcPy30x/q6sFo/NinNB5zaMs50R8rLoLFVE0 Dnbgdgd1z7FEMA34wcY4ThlJIxYBNeicN5wo5QE= X-Google-Smtp-Source: AAOMgpezAp+bs/9Za7En6ah4QM+XTSYNU8dkm4iU0LKXF5lIf1jQFrYIomPd2l1Wx5Tm3701OC3qbLOfA2C2kWSUhkI= X-Received: by 2002:a37:21e6:: with SMTP id f99-v6mr9630518qki.206.1532795787033; Sat, 28 Jul 2018 09:36:27 -0700 (PDT) MIME-Version: 1.0 References: <1357421162.31089.1531812291583@webmail.strato.de> <1c323544b3076c0ab31b887d6113f25f572e41ae.camel@coverfire.com> <3ab0c2e8d33a71ca58a404a6a8b293a95fe1e9d3.camel@coverfire.com> <33C09185-34D2-4D96-9DE3-345D51D4D5C9@gmail.com> In-Reply-To: <33C09185-34D2-4D96-9DE3-345D51D4D5C9@gmail.com> From: Dave Taht Date: Sat, 28 Jul 2018 09:36:47 -0700 Message-ID: To: Jonathan Morton Cc: Dan Siemon , Cake List Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Subject: Re: [Cake] =?utf-8?q?Using_cake_to_shape_1000=E2=80=99s_of_users=2E?= X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 28 Jul 2018 16:36:27 -0000 On Sat, Jul 28, 2018 at 9:11 AM Jonathan Morton wro= te: > > > On 28 Jul, 2018, at 6:51 pm, Dave Taht wrote: > > > > That's also pretty low end. On the high end nowadays there's stuff like= this: > > > > https://www.amazon.com/Intel-Xeon-E5-2698-Hexadeca-core-Processor/dp/B0= 0PDD1QES > > Intel is no longer high-end for x86 CPUs. Not all of the market has real= ised that yet, but it's true. > > Look at Threadripper 2 which scales up to 32 cores, 64 threads in a singl= e socket at HEDT prices, and EPYC which just goes bonkers in terms of I/O c= apabilities and still costs less than its nearest Intel competitor. None o= f which has any serious concerns with the recent series of Meltdown/Spectre= speculation bugs, unlike Intel. That haunts me. > AMD is moving to a 7nm silicon process which apparently works pretty well= already, and is theoretically on par with Intel's 10nm process which they = still haven't got working reliably after how many years of delays now? And= they're already beating Intel over the head with a 14nm process which is t= heoretically *inferior* to Intel's 14nm process, which they'll be stuck wit= h in practice for *at least* the next year even by their own wildly optimis= tic latest estimates. > > The only place Intel temporarily holds a real advantage is in maximum sin= gle-threaded turbo clock speed. Which is important. Myself, I dream of cpus that can context switch in 5 cycles like the mill. For I/O heavy workloads it's context switching that is rather important. There is also a new generation of multi-core arms, some with devkits like netronome's and caviums. My take on all this would be to try and benchmark some stuff. Some other bottleneck will surface. We have one benchmark of 4000 users at 4GigE to work with at the moment. I've long pointed out that one big reason for GRO/GSO is the cost of a routing lookup. mellonox, I think(?), *finally* added TCAM support to their cards in a patch that just went by on the netdev mailing list. > This is relevant to a shrinking minority of users these days. AMD's next= CPUs are supposed to make that wholly irrelevant with a significant furthe= r jump in IPC - because they were designed to compete with 10nm Intel CPUs = that Intel now looks very unlikely to be capable of manufacturing. One huge advantage intel had was dma into cache. Still? > > And is this relevant to "Super Mega Turbo Cake Edition XLRi"? Hahahaha. That comes up with 7000 hits for a hair dryer. https://www.google.com/search?q=3DSuper+Mega+Turbo+Cake+Edition+XLRi&rlz=3D= 1C5CHFA_enUS749US749&oq=3DSuper+Mega+Turbo+Cake+Edition+XLRi&aqs=3Dchrome..= 69i57.381j0j4&sourceid=3Dchrome&ie=3DUTF-8 > Well, one of the nice things about having lots of users is that you can s= tatistically multiplex them across multiple hardware queues more easily. E= ach subscriber's traffic can sanely end up on the same queue each time, and= each queue can have a separate "Super Cake" instance allocated an even div= ision of the total backhaul bandwidth, and in theory each of *those* can ru= n on its own CPU core. Instant throughput boost. I'd believe it when I benchmarked it, not before. > > - Jonathan Morton > --=20 Dave T=C3=A4ht CEO, TekLibre, LLC http://www.teklibre.com Tel: 1-669-226-2619