From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <brunogm0@gmail.com>
Received: from mail-yw1-xc32.google.com (mail-yw1-xc32.google.com
 [IPv6:2607:f8b0:4864:20::c32])
 (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 B38763B2A4
 for <bloat@lists.bufferbloat.net>; Wed, 28 Nov 2018 11:32:22 -0500 (EST)
Received: by mail-yw1-xc32.google.com with SMTP id h138-v6so9140700ywa.11
 for <bloat@lists.bufferbloat.net>; Wed, 28 Nov 2018 08:32:22 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;
 h=mime-version:from:date:message-id:subject:to;
 bh=Kq5BFE2edH5Dgsm2ar9Xg5eJPWong7k3BfO+YDOtgrw=;
 b=Zflh1MJXi8+kbgEtNrSm8E9HVwHRHtf12dOGi/JbC19zWA2M332oO+Hn2ebs4A3bHY
 r2XfPw9S1j8eWbP88v5yDCHFqfLOnKau6sL2voMJPVxaIy4P/8ezEqI4zJwBOx9658Te
 YIlM2d3UoeCyU8U4lNKVXhwoQdx9oatf27TK92TwklSsANCfTx6p4eQeOBxS2eSmUpGd
 BpnSdwm2UeKjczAuuxUKlcKXll8NgLMY2lXE/MZDhcFm3/bttlfLweKjQPZ2QLZr069n
 PFsEwdWqyzUjR/VJ8h9ED5ftjb8sv1Hwc8abopiHJRgrwUYQSYYdaNCvHqWC3GWLV/T1
 Lvsw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
 d=1e100.net; s=20161025;
 h=x-gm-message-state:mime-version:from:date:message-id:subject:to;
 bh=Kq5BFE2edH5Dgsm2ar9Xg5eJPWong7k3BfO+YDOtgrw=;
 b=nnG4ep929GnHFkJhly9sZKHC1jtU4XXERbWlqyKmf/8ZHWlmMlEDgvu4I6XlxLnbKp
 xyeqvpylaCTXJWfV2BoelRPP1Y8CscXC0okesOJbJo/T+lqQucSPI0Vf9DMPX4cAZuoH
 JKmRtE8bZXtNM1/0kax6a2mwpdo1EbLBMddw/sOWG7BbOhoGe09ZOfo3OBNLYBzVCiEh
 6vdEw5YbKIB8mzfezwhmphp07BkPEMWJoy0rxPe4oQ76Oa9CXgMPTWYdprcdsI6BCI9s
 IBMlD0RaUSPhEUll37HK2er0enxxBsGxPmjoICbyIYcaEJHFaDz3GoJcH+Czq5srtEQp
 Jwdw==
X-Gm-Message-State: AGRZ1gJlB0deo04mv6OVBxtxrnTBuihvrCfvlEtWt+Q8GO9lWj4/82VT
 v5v77ieGgk5Npisg5alZPoSdkmNGwMw6iM+PAr7Zw0MA
X-Google-Smtp-Source: AJdET5cnkyAwZKZiV5hqlkyBRA26HC56FFUE+KbE/Z+ccJTbNdKQjDb73qFHtZkwruESGc9TSlsR2137emcjJSG8tMU=
X-Received: by 2002:a0d:e887:: with SMTP id r129mr39835776ywe.54.1543422741985; 
 Wed, 28 Nov 2018 08:32:21 -0800 (PST)
MIME-Version: 1.0
From: Bruno George Moraes <brunogm0@gmail.com>
Date: Wed, 28 Nov 2018 14:32:10 -0200
Message-ID: <CAJpr3FMOr+zC1_oEnDU0s+67b2SLPSELo4esLQY-UCJtOGtKNw@mail.gmail.com>
To: Bufferbloat lists bloat <bloat@lists.bufferbloat.net>
Content-Type: multipart/alternative; boundary="000000000000463d23057bbc1cdd"
Subject: [Bloat]  known buffer sizes on switches
X-BeenThere: bloat@lists.bufferbloat.net
X-Mailman-Version: 2.1.20
Precedence: list
List-Id: General list for discussing Bufferbloat <bloat.lists.bufferbloat.net>
List-Unsubscribe: <https://lists.bufferbloat.net/options/bloat>,
 <mailto:bloat-request@lists.bufferbloat.net?subject=unsubscribe>
List-Archive: <https://lists.bufferbloat.net/pipermail/bloat>
List-Post: <mailto:bloat@lists.bufferbloat.net>
List-Help: <mailto:bloat-request@lists.bufferbloat.net?subject=help>
List-Subscribe: <https://lists.bufferbloat.net/listinfo/bloat>,
 <mailto:bloat-request@lists.bufferbloat.net?subject=subscribe>
X-List-Received-Date: Wed, 28 Nov 2018 16:32:22 -0000

--000000000000463d23057bbc1cdd
Content-Type: text/plain; charset="UTF-8"

>
> Nice resource, thanks.
>
> If someone wonders why things look the way they do, so it's all about
> on-die and off-die memory. Either you use off-die or on-die memory, often
> SRAM which requires 6 gates per bit. So spending half a billion gates
> gives you ~10MB buffer on-die. If you're doing off-die memory (DRAM or
> similar) then you'll get the gigabytes of memory seen in some equipment.
> There basically is nothing in between. As soon as you go off-die you might
> as well put at least 2-6 GB in there.
>
>
There are some reasearch on new memory devices with unexpected results...
https://ieeexplore.ieee.org/document/8533260

The HMC memory allows improvements in execution time and consumed energy.
> In some situations, this memory type permits removing the L2 cache from the
> memory hierarchy.
>

HMC parts start at 2GB

--000000000000463d23057bbc1cdd
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><blockquote class=3D"gmail_quote" style=
=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding=
-left:1ex">
<pre>Nice resource, thanks.

If someone wonders why things look the way they do, so it&#39;s all about=
=20
on-die and off-die memory. Either you use off-die or on-die memory, often=
=20
SRAM which requires 6 gates per bit. So spending half a billion gates=20
gives you ~10MB buffer on-die. If you&#39;re doing off-die memory (DRAM or=
=20
similar) then you&#39;ll get the gigabytes of memory seen in some equipment=
.=20
There basically is nothing in between. As soon as you go off-die you might=
=20
as well put at least 2-6 GB in there.
</pre></blockquote><div><br></div><div>There are some reasearch on new memo=
ry devices with unexpected results...</div><div><a href=3D"https://ieeexplo=
re.ieee.org/document/8533260">https://ieeexplore.ieee.org/document/8533260<=
/a></div><div><br></div><blockquote class=3D"gmail_quote" style=3D"margin:0=
px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><=
div>
The HMC memory allows improvements in execution time and consumed=20
energy. In some situations, this memory type permits removing the L2=20
cache from the memory hierarchy.=20

</div></blockquote><div><br></div><div>HMC parts start at 2GB <br></div><di=
v><br></div><div><br></div></div></div>

--000000000000463d23057bbc1cdd--