Return-Path: X-Original-To: quic-issues@ietfa.amsl.com Delivered-To: quic-issues@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F1A5A3A0061 for ; Thu, 14 May 2020 00:03:34 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.273 X-Spam-Level: X-Spam-Status: No, score=-3.273 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.173, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, MAILING_LIST_MULTI=-1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=github.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xvLQIGOagOEY for ; Thu, 14 May 2020 00:03:33 -0700 (PDT) Received: from out-28.smtp.github.com (out-28.smtp.github.com [192.30.252.211]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 827E73A00AD for ; Thu, 14 May 2020 00:03:33 -0700 (PDT) Received: from github-lowworker-6b40fdd.va3-iad.github.net (github-lowworker-6b40fdd.va3-iad.github.net [10.48.16.64]) by smtp.github.com (Postfix) with ESMTP id 57A0F8C0460 for ; Thu, 14 May 2020 00:03:32 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=github.com; s=pf2014; t=1589439812; bh=1w9V3BFvHqwI5kDlZOZV8BUE1iRlxB0FNe27Jcdf9hY=; h=Date:From:Reply-To:To:Cc:In-Reply-To:References:Subject:List-ID: List-Archive:List-Post:List-Unsubscribe:From; b=M5d0m+yGNgmdapcBKtZ+bOapNPha9N2Z4cFv+FQFbBsRKddIzMQbPXWEgURT4PtrI ko/oOwUdKRyMOvzsYDvCKkET1eQ5j0u9y0I3O/m4BpIH1TVgt+IZpr2sC6BTdPZl4c scbV2VWvRpHMxQ/4WSmlM5+o+wfWPgxtNI+vOKro= Date: Thu, 14 May 2020 00:03:32 -0700 From: Jana Iyengar Reply-To: quicwg/base-drafts To: quicwg/base-drafts Cc: Subscribed Message-ID: In-Reply-To: References: Subject: Re: [quicwg/base-drafts] A pacing algorithm (#3630) Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="--==_mimepart_5ebced44484f1_3cb83fafbfccd964751f9"; charset=UTF-8 Content-Transfer-Encoding: 7bit Precedence: list X-GitHub-Sender: janaiyengar X-GitHub-Recipient: quic-issues X-GitHub-Reason: subscribed X-Auto-Response-Suppress: All X-GitHub-Recipient-Address: quic-issues@ietf.org Archived-At: X-BeenThere: quic-issues@ietf.org X-Mailman-Version: 2.1.29 List-Id: Notification list for GitHub issues related to the QUIC WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 May 2020 07:03:35 -0000 ----==_mimepart_5ebced44484f1_3cb83fafbfccd964751f9 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit @janaiyengar commented on this pull request. > +as the one in this document, that rate can be computed by averaging the +congestion window over the round-trip time. Expressed as a rate in bytes: + +~~~ +rate = N * congestion_window / smoothed_rtt +~~~ + +Or, expressed as an inter-packet interval: + +~~~ +interval = smoothed_rtt * packet_size / congestion_window / N +~~~ + +Using a value for N that is small, but greater than 1 (for example, 1.25) +ensures that short-term variations in round-trip time or scheduler delays don't +result in not fully utilizing the congestion window. ```suggestion result in under-utilization of the congestion window. ``` > +as the one in this document, that rate can be computed by averaging the +congestion window over the round-trip time. Expressed as a rate in bytes: + +~~~ +rate = N * congestion_window / smoothed_rtt +~~~ + +Or, expressed as an inter-packet interval: + +~~~ +interval = smoothed_rtt * packet_size / congestion_window / N +~~~ + +Using a value for `N` that is small, but at least 1 (for example, 1.25) ensures +that short-term variations in round-trip time or scheduler delays don't result +in not fully utilizing the congestion window. ```suggestion in under-utilization of the congestion window. ``` > @@ -897,9 +897,28 @@ congestion window, which is recommended to be the minimum of max_datagram_size is the current maximum size of a datagram for the connection, not including UDP or IP overhead. -As an example of a well-known and publicly available implementation of a flow -pacer, implementers are referred to the Fair Queue packet scheduler (fq qdisc) -in Linux (3.11 onwards). +Endpoints can implement pacing as they choose. A perfectly paced sender spreads I would propose moving this paragraph up one, and adding this to the start of the previous paragraph: "Practical considerations, such as packetization, scheduling delays, and computational efficiency, can cause a sender to deviate from this rate over time periods that are much shorter than a round-trip time. Sending multiple packets into the network without any delay ..." -- 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/pull/3630#pullrequestreview-411514042 ----==_mimepart_5ebced44484f1_3cb83fafbfccd964751f9 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

=0D

@janaiyengar commented on this pull request.

=0D =0D
=0D =0D

In draft-ietf-quic-recovery.md:

=0D 
> +as the one in this document, that rate ca=
n be computed by averaging the=0D
+congestion window over the round-trip time. Expressed as a rate in bytes=
:=0D
+=0D
+~~~=0D
+rate =3D N * congestion_window / smoothed_rtt=0D
+~~~=0D
+=0D
+Or, expressed as an inter-packet interval:=0D
+=0D
+~~~=0D
+interval =3D smoothed_rtt * packet_size / congestion_window / N=0D
+~~~=0D
+=0D
+Using a value for N that is small, but greater than 1 (for example, 1.25=
)=0D
+ensures that short-term variations in round-trip time or scheduler delay=
s don't=0D
+result in not fully utilizing the congestion window.=0D
=0D =E2=AC=87=EF=B8=8F Suggested change=0D 
-result in not fully utilizing the congestion =
window.=0D
+result in under-utilization of the congestion window.=0D
=0D =0D =0D
=0D =0D

In draft-ietf-quic-recovery.md:

=0D 
> +as the one in this document, that rate ca=
n be computed by averaging the=0D
+congestion window over the round-trip time. Expressed as a rate in bytes=
:=0D
+=0D
+~~~=0D
+rate =3D N * congestion_window / smoothed_rtt=0D
+~~~=0D
+=0D
+Or, expressed as an inter-packet interval:=0D
+=0D
+~~~=0D
+interval =3D smoothed_rtt * packet_size / congestion_window / N=0D
+~~~=0D
+=0D
+Using a value for `N` that is small, but at least 1 (for example, 1.25) =
ensures=0D
+that short-term variations in round-trip time or scheduler delays don=
9;t result=0D
+in not fully utilizing the congestion window.=0D
=0D =E2=AC=87=EF=B8=8F Suggested change=0D 
-in not fully utilizing the congestion window.=
=0D
+in under-utilization of the congestion window.=0D
=0D =0D =0D
=0D =0D

In draft-ietf-quic-recovery.md:

=0D 
> @@ -897,9 +897,28 @@ congestion window, wh=
ich is recommended to be the minimum of=0D
 max_datagram_size is the current maximum size of a datagram for the conn=
ection,=0D
 not including UDP or IP overhead.=0D
 =0D
-As an example of a well-known and publicly available implementation of a=
 flow=0D
-pacer, implementers are referred to the Fair Queue packet scheduler (fq =
qdisc)=0D
-in Linux (3.11 onwards).=0D
+Endpoints can implement pacing as they choose. A perfectly paced sender =
spreads=0D
=0D

I would propose moving this paragraph up one, and adding this to the s= tart of the previous paragraph: "Practical considerations, such as packet= ization, scheduling delays, and computational efficiency, can cause a sen= der to deviate from this rate over time periods that are much shorter tha= n a round-trip time. Sending multiple packets into the network without an= y delay ..."

=0D =0D

&m= dash;
You are receiving this because you are subscribed to this thre= ad.
Reply to this email directly, view it on GitHub, or unsubscribe.3D""

=0D = ----==_mimepart_5ebced44484f1_3cb83fafbfccd964751f9--