Date: Sun, 12 May 2019 18:03:36 -0700
From: ianswett <notifications@github.com>
Reply-To: quicwg/base-drafts
 <reply+AFTOJK6POAINCZRCPJZPVMF24X2OREVBNHHBTZE2DQ@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/issues/2630/491645245@github.com>
In-Reply-To: <quicwg/base-drafts/issues/2630@github.com>
References: <quicwg/base-drafts/issues/2630@github.com>
Subject: Re: [quicwg/base-drafts] Define "under-utilization" of cwnd (#2630)
Mime-Version: 1.0
Content-Type: multipart/alternative;
 boundary="--==_mimepart_5cd8c26882ac3_28503fe6f2ccd968715886";
 charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: list
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/04MZtXvmnofLSSez1lPddOIQnFE>


----==_mimepart_5cd8c26882ac3_28503fe6f2ccd968715886
Content-Type: text/plain;
 charset=UTF-8
Content-Transfer-Encoding: 7bit

@janaiyengar Good point about the in_flight > 1/2 CWND check in Linux.  In BBR, that only applies to the first doubling of IW, and after that it's fully bandwidth based, FWIW.

We could add a similar constraint that this rule can be applied when the window was not fully used at the time of sending, but was at the time of acknowledgement as long as the window is < 2*IW?

One clear goal of mine is I don't want the handshake alone increasing the congestion window(assuming IW10 and a moderately sized handshake).

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/2630#issuecomment-491645245
----==_mimepart_5cd8c26882ac3_28503fe6f2ccd968715886
Content-Type: text/html;
 charset=UTF-8
Content-Transfer-Encoding: 7bit

<p><a class="user-mention" data-hovercard-type="user" data-hovercard-url="/hovercards?user_id=11067604" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/janaiyengar">@janaiyengar</a> Good point about the in_flight &gt; 1/2 CWND check in Linux.  In BBR, that only applies to the first doubling of IW, and after that it's fully bandwidth based, FWIW.</p>
<p>We could add a similar constraint that this rule can be applied when the window was not fully used at the time of sending, but was at the time of acknowledgement as long as the window is &lt; 2*IW?</p>
<p>One clear goal of mine is I don't want the handshake alone increasing the congestion window(assuming IW10 and a moderately sized handshake).</p>

<p style="font-size:small;-webkit-text-size-adjust:none;color:#666;">&mdash;<br />You are receiving this because you are subscribed to this thread.<br />Reply to this email directly, <a href="https://github.com/quicwg/base-drafts/issues/2630#issuecomment-491645245">view it on GitHub</a>, or <a href="https://github.com/notifications/unsubscribe-auth/AFTOJK4ZTWV3YRE6FHH2W63PVC46RANCNFSM4HGWQM2A">mute the thread</a>.<img src="https://github.com/notifications/beacon/AFTOJK3HPTXMDMVNXSTA66DPVC46RA5CNFSM4HGWQM2KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVG6SPI.gif" height="1" width="1" alt="" /></p>
<script type="application/ld+json">[
{
"@context": "http://schema.org",
"@type": "EmailMessage",
"potentialAction": {
"@type": "ViewAction",
"target": "https://github.com/quicwg/base-drafts/issues/2630#issuecomment-491645245",
"url": "https://github.com/quicwg/base-drafts/issues/2630#issuecomment-491645245",
"name": "View Issue"
},
"description": "View this Issue on GitHub",
"publisher": {
"@type": "Organization",
"name": "GitHub",
"url": "https://github.com"
}
}
]</script>
----==_mimepart_5cd8c26882ac3_28503fe6f2ccd968715886--