<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p>Matthias, Dave,<br>
</p>
<p>The sort of industrial control applications that detnet is
targeting require far lower queuing delay and jitter than fq_CoDel
can give. They have thrown around numbers like 250us jitter and
1E-9 to 1E-12 packet loss probability.</p>
<p>However, like you, I just sigh when I see the behemoth detnet is
building.</p>
<p>Nonetheless, it's important to have a debate about where to go to
next. Personally I don't think fq_CoDel alone has legs to get
(that) much better. <br>
</p>
<p> I prefer the direction that Mohamad Alizadeh's HULL pointed in:
<br>
<a moz-do-not-send="true"
href="https://people.csail.mit.edu/alizadeh/papers/hull-nsdi12.pdf">Less
is More: Trading a little Bandwidth for Ultra-Low Latency in the
Data Center</a><br>
</p>
<p>In HULL you have i) a virtual queue that models what the queue
would be if the link were slightly slower, then marks with ECN
based on that. ii) a much more well-behaved TCP (HULL uses DCTCP
with hardware pacing in the NICs). <br>
</p>
<p>I would love to be able to demonstrate that HULL can achieve the
same extremely low latency and loss targets as detnet, but with a
fraction of the complexity.</p>
<p><b>Queuing latency?</b> This keeps the real FIFO queue in the low
hundreds to tens of microseconds. </p>
<p><b>Loss prob?</b> Mohammad doesn't recall seeing a loss during
the entire period of the experiments, but he doubted their
measurement infrastructure was sufficiently accurate (or went on
long enough) to be sure they were able to detect one loss per
10^12 packets.<br>
</p>
<p>For their research prototype, HULL used a dongle they built,
plugged into each output port to constrict the link in order to
shift the AQM out of the box. However, Broadcom mid-range chipsets
already contain vertual queue hardware (courtesey of a project we
did with them when I was at BT:<br>
<a moz-do-not-send="true"
href="http://bobbriscoe.net/pubs.html#vq2lb">How to Build a
Virtual Queue from Two Leaky Buckets (and why one is not enough)</a>
).<br>
</p>
<b>For public Internet, not just for DCs?</b> You might have seen
the work we've done (<a moz-do-not-send="true"
href="https://riteproject.eu/dctth/">L4S</a>) to get queuing delay
over regular public Internet and broadband down to about mean 500us;
90%-ile 1ms, by making DCTCP deployable alongside existing Internet
traffic (unlike HULL, pacing at the source is in Linux, not
hardware). My personal roadmap for that is to introduce virtual
queues at some future stage, to get down to the sort of delays that
detnet wants, but over the public Internet with just FIFOs. <br>
<br>
Radio links are harder, of course, but a lot of us are working on
that too.<br>
<br>
<br>
<br>
Bob<br>
<br>
<div class="moz-cite-prefix">On 12/11/2017 22:58, Matthias
Tafelmeier wrote:<br>
</div>
<blockquote type="cite"
cite="mid:79f4d92c-74f4-8cd0-9d38-e51a668cb9b6@gmx.net">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<div class="moz-cite-prefix">On 11/07/2017 01:36 AM, Dave Taht
wrote:<br>
</div>
<blockquote type="cite" cite="mid:87shdr0vt6.fsf@nemesis.taht.net">
<blockquote type="cite" style="color: #000000;">
<pre wrap="">Perceived that as shareworthy/entertaining ..
<a class="moz-txt-link-freetext" href="https://tools.ietf.org/html/draft-ietf-detnet-architecture-03#section-4.5" moz-do-not-send="true">https://tools.ietf.org/html/draft-ietf-detnet-architecture-03#section-4.5</a>
without wanting to belittle it.
</pre>
</blockquote>
<pre wrap="">Hope springs eternal that they might want to look over the relevant
codel and fq_codel RFCS at some point or another.
</pre>
</blockquote>
<p>Not sure, appears like juxtaposing classical mechanics to
nanoscale physics.<br>
</p>
<pre class="moz-signature" cols="72">--
Besten Gruß
Matthias Tafelmeier
</pre>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Bloat mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Bloat@lists.bufferbloat.net">Bloat@lists.bufferbloat.net</a>
<a class="moz-txt-link-freetext" href="https://lists.bufferbloat.net/listinfo/bloat">https://lists.bufferbloat.net/listinfo/bloat</a>
</pre>
</blockquote>
<br>
</body>
</html>