< draft-ietf-tcpm-dctcp-02.txt   draft-ietf-tcpm-dctcp-03.txt >
Network Working Group S. Bensley Network Working Group S. Bensley
Internet-Draft Microsoft Internet-Draft Microsoft
Intended status: Informational L. Eggert Intended status: Informational L. Eggert
Expires: January 19, 2017 NetApp Expires: May 18, 2017 NetApp
D. Thaler D. Thaler
P. Balasubramanian P. Balasubramanian
Microsoft Microsoft
G. Judd G. Judd
Morgan Stanley Morgan Stanley
July 18, 2016 November 14, 2016
Datacenter TCP (DCTCP): TCP Congestion Control for Datacenters Datacenter TCP (DCTCP): TCP Congestion Control for Datacenters
draft-ietf-tcpm-dctcp-02 draft-ietf-tcpm-dctcp-03
Abstract Abstract
This informational memo describes Datacenter TCP (DCTCP), an This informational memo describes Datacenter TCP (DCTCP), an
improvement to TCP congestion control for datacenter traffic. DCTCP improvement to TCP congestion control for datacenter traffic. DCTCP
uses improved Explicit Congestion Notification (ECN) processing to uses improved Explicit Congestion Notification (ECN) processing to
estimate the fraction of bytes that encounter congestion, rather than estimate the fraction of bytes that encounter congestion, rather than
simply detecting that some congestion has occurred. DCTCP then simply detecting that some congestion has occurred. DCTCP then
scales the TCP congestion window based on this estimate. This method scales the TCP congestion window based on this estimate. This method
achieves high burst tolerance, low latency, and high throughput with achieves high burst tolerance, low latency, and high throughput with
shallow-buffered switches. This memo also discusses deployment shallow-buffered switches. This memo also discusses deployment
issues related to the coexistence of DCTCP and conventional TCP, the issues related to the coexistence of DCTCP and conventional TCP, the
lack of a negotiating mechanism between sender and receiver, and lack of a negotiating mechanism between sender and receiver, and
presents some possible mitigations. presents some possible mitigations. DCTCP as described in this draft
is applicable to deployments in controlled environments like
datacenters but it MUST NOT be deployed over the public Internet
without additional measures, as detailed in Section 5.
Status of This Memo Status of This Memo
This Internet-Draft is submitted in full conformance with the This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79. provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet- working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/. Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
This Internet-Draft will expire on January 19, 2017. This Internet-Draft will expire on May 18, 2017.
Copyright Notice Copyright Notice
Copyright (c) 2016 IETF Trust and the persons identified as the Copyright (c) 2016 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of (http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
skipping to change at page 2, line 25 skipping to change at page 2, line 25
to this document. Code Components extracted from this document must to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License. described in the Simplified BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. DCTCP Algorithm . . . . . . . . . . . . . . . . . . . . . . . 4 3. DCTCP Algorithm . . . . . . . . . . . . . . . . . . . . . . . 4
3.1. Marking Congestion on the Switches . . . . . . . . . . . 4 3.1. Marking Congestion on the L3 Switches and Routers . . . . 4
3.2. Echoing Congestion Information on the Receiver . . . . . 4 3.2. Echoing Congestion Information on the Receiver . . . . . 4
3.3. Processing Congestion Indications on the Sender . . . . . 5 3.3. Processing Congestion Indications on the Sender . . . . . 6
3.4. Handling of SYN, SYN-ACK, RST Packets . . . . . . . . . . 7 3.4. Handling of SYN, SYN-ACK, RST Packets . . . . . . . . . . 8
4. Implementation Issues . . . . . . . . . . . . . . . . . . . . 8 4. Implementation Issues . . . . . . . . . . . . . . . . . . . . 8
5. Deployment Issues . . . . . . . . . . . . . . . . . . . . . . 9 5. Deployment Issues . . . . . . . . . . . . . . . . . . . . . . 9
6. Known Issues . . . . . . . . . . . . . . . . . . . . . . . . 10 6. Known Issues . . . . . . . . . . . . . . . . . . . . . . . . 10
7. Implementation Status . . . . . . . . . . . . . . . . . . . . 11 7. Implementation Status . . . . . . . . . . . . . . . . . . . . 11
8. Security Considerations . . . . . . . . . . . . . . . . . . . 11 8. Security Considerations . . . . . . . . . . . . . . . . . . . 11
9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 11 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 12
10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 11 10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 12
11. References . . . . . . . . . . . . . . . . . . . . . . . . . 12 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 12
11.1. Normative References . . . . . . . . . . . . . . . . . . 12 11.1. Normative References . . . . . . . . . . . . . . . . . . 12
11.2. Informative References . . . . . . . . . . . . . . . . . 12 11.2. Informative References . . . . . . . . . . . . . . . . . 13
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 13 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14
1. Introduction 1. Introduction
Large datacenters necessarily need many network switches to Large datacenters necessarily need many network switches to
interconnect their many servers. Therefore, a datacenter can greatly interconnect their many servers. Therefore, a datacenter can greatly
reduce its capital expenditure by leveraging low-cost switches. reduce its capital expenditure by leveraging low-cost switches.
However, such low-cost switches tend to have limited queue capacities However, such low-cost switches tend to have limited queue capacities
and are thus more susceptible to packet loss due to congestion. and are thus more susceptible to packet loss due to congestion.
Network traffic in a datacenter is often a mix of short and long Network traffic in a datacenter is often a mix of short and long
skipping to change at page 3, line 18 skipping to change at page 3, line 18
These factors place some conflicting demands on the queue occupancy These factors place some conflicting demands on the queue occupancy
of a switch: of a switch:
o The queue must be short enough that it does not impose excessive o The queue must be short enough that it does not impose excessive
latency on short flows. latency on short flows.
o The queue must be long enough to buffer sufficient data for the o The queue must be long enough to buffer sufficient data for the
long flows to saturate the path capacity. long flows to saturate the path capacity.
o The queue must be short enough to absorb incast bursts without o The queue must be long enough to absorb incast bursts without
excessive packet loss. excessive packet loss.
Standard TCP congestion control [RFC5681] relies on packet loss to Standard TCP congestion control [RFC5681] relies on packet loss to
detect congestion. This does not meet the demands described above. detect congestion. This does not meet the demands described above.
First, short flows will start to experience unacceptable latencies First, short flows will start to experience unacceptable latencies
before packet loss occurs. Second, by the time TCP congestion before packet loss occurs. Second, by the time TCP congestion
control kicks in on the senders, most of the incast burst has already control kicks in on the senders, most of the incast burst has already
been dropped. been dropped.
[RFC3168] describes a mechanism for using Explicit Congestion [RFC3168] describes a mechanism for using Explicit Congestion
skipping to change at page 3, line 45 skipping to change at page 3, line 45
Datacenter TCP (DCTCP) improves traditional ECN processing by Datacenter TCP (DCTCP) improves traditional ECN processing by
estimating the fraction of bytes that encounter congestion, rather estimating the fraction of bytes that encounter congestion, rather
than simply detecting that some congestion has occurred. DCTCP then than simply detecting that some congestion has occurred. DCTCP then
scales the TCP congestion window based on this estimate. This method scales the TCP congestion window based on this estimate. This method
achieves high burst tolerance, low latency, and high throughput with achieves high burst tolerance, low latency, and high throughput with
shallow-buffered switches. shallow-buffered switches.
It is recommended that DCTCP be only deployed in a datacenter It is recommended that DCTCP be only deployed in a datacenter
environment where the endpoints and the switching fabric are under a environment where the endpoints and the switching fabric are under a
single administrative domain. This protocol is not meant for single administrative domain. This protocol is not meant for
uncontrolled deployment in the global Internet. uncontrolled deployment in the global Internet. Refer to Section 5
for more details.
2. Terminology 2. Terminology
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in [RFC2119]. Normative document are to be interpreted as described in [RFC2119]. Normative
language is used to describe how necessary the various aspects of the language is used to describe how necessary the various aspects of the
Microsoft implementation are for interoperability, but even compliant Microsoft implementation are for interoperability, but even compliant
implementations without the measures in sections 4-6 would still only implementations without the measures in sections 4-6 would still only
be safe to deploy in controlled environments. be safe to deploy in controlled environments.
skipping to change at page 4, line 22 skipping to change at page 4, line 23
o The switches (or other intermediate devices in the network) detect o The switches (or other intermediate devices in the network) detect
congestion and set the Congestion Encountered (CE) codepoint in congestion and set the Congestion Encountered (CE) codepoint in
the IP header. the IP header.
o The receiver echoes the congestion information back to the sender, o The receiver echoes the congestion information back to the sender,
using the ECN-Echo (ECE) flag in the TCP header. using the ECN-Echo (ECE) flag in the TCP header.
o The sender computes a congestion estimate and reacts, by reducing o The sender computes a congestion estimate and reacts, by reducing
the TCP congestion window accordingly (cwnd). the TCP congestion window accordingly (cwnd).
3.1. Marking Congestion on the Switches 3.1. Marking Congestion on the L3 Switches and Routers
The switches in a datacenter fabric indicate congestion to the end The L3 switches and routers in a datacenter fabric indicate
nodes by setting the CE codepoint in the IP header as specified in congestion to the end nodes by setting the CE codepoint in the IP
Section 5 of [RFC3168]. For example, the switches may be configured header as specified in Section 5 of [RFC3168]. For example, the
with a congestion threshold. When a packet arrives at a switch and switches may be configured with a congestion threshold. When a
its queue length is greater than the congestion threshold, the switch packet arrives at a switch and its queue length is greater than the
sets the CE codepoint in the packet. For example, Section 3.4 of congestion threshold, the switch sets the CE codepoint in the packet.
[DCTCP10] suggests threshold marking with a threshold K > (RTT * For example, Section 3.4 of [DCTCP10] suggests threshold marking with
C)/7, where C is the link rate in packets per second. However, the a threshold K > (RTT * C)/7, where C is the link rate in packets per
actual algorithm for marking congestion is an implementation detail second. However, the actual algorithm for marking congestion is an
of the switch and will generally not be known to the sender and implementation detail of the switch and will generally not be known
receiver. Therefore, sender and receiver should not assume that a to the sender and receiver. Therefore, sender and receiver should
particular marking algorithm is implemented by the switching fabric. not assume that a particular marking algorithm is implemented by the
switching fabric.
3.2. Echoing Congestion Information on the Receiver 3.2. Echoing Congestion Information on the Receiver
According to Section 6.1.3 of [RFC3168], the receiver sets the ECE According to Section 6.1.3 of [RFC3168], the receiver sets the ECE
flag if any of the packets being acknowledged had the CE code point flag if any of the packets being acknowledged had the CE code point
set. The receiver then continues to set the ECE flag until it set. The receiver then continues to set the ECE flag until it
receives a packet with the Congestion Window Reduced (CWR) flag set. receives a packet with the Congestion Window Reduced (CWR) flag set.
However, the DCTCP algorithm requires more detailed congestion However, the DCTCP algorithm requires more detailed congestion
information. In particular, the sender must be able to determine the information. In particular, the sender must be able to determine the
number of bytes sent that encountered congestion. Thus, the scheme number of bytes sent that encountered congestion. Thus, the scheme
described in [RFC3168] does not suffice. described in [RFC3168] does not suffice.
One possible solution is to ACK every packet and set the ECE flag in One possible solution is to ACK every packet and set the ECE flag in
the ACK if and only if the CE code point was set in the packet being the ACK if and only if the CE code point was set in the packet being
acknowledged. However, this prevents the use of delayed ACKs, which acknowledged. However, this prevents the use of delayed ACKs, which
are an important performance optimization in datacenters. are an important performance optimization in datacenters. If the
delayed ACK frequency is m, then an ACK is generated every m packets.
The typical value of m is 2 but it could be affected by ACK
throttling or packet coalescing techniques designed to improve
performance.
Instead, DCTCP introduces a new Boolean TCP state variable, "DCTCP Instead, DCTCP introduces a new Boolean TCP state variable, "DCTCP
Congestion Encountered" (DCTCP.CE), which is initialized to false and Congestion Encountered" (DCTCP.CE), which is initialized to false and
stored in the Transmission Control Block (TCB). When sending an ACK, stored in the Transmission Control Block (TCB). When sending an ACK,
the ECE flag MUST be set if and only if DCTCP.CE is true. When the ECE flag MUST be set if and only if DCTCP.CE is true. When
receiving packets, the CE codepoint MUST be processed as follows: receiving packets, the CE codepoint MUST be processed as follows:
1. If the CE codepoint is set and DCTCP.CE is false, send an ACK for 1. If the CE codepoint is set and DCTCP.CE is false, send an ACK for
any previously unacknowledged packets and set DCTCP.CE to true. any previously unacknowledged packets and set DCTCP.CE to true.
2. If the CE codepoint is not set and DCTCP.CE is true, send an ACK 2. If the CE codepoint is not set and DCTCP.CE is true, send an ACK
for any previously unacknowledged packets and set DCTCP.CE to for any previously unacknowledged packets and set DCTCP.CE to
false. false.
3. Otherwise, ignore the CE codepoint. 3. Otherwise, ignore the CE codepoint.
The immediate ACK generated SHOULD NOT acknowledge any data in the
received packet that changes the DCTCP.CE state.
Receiver handling of the "Congestion Window Reduced" (CWR) bit is Receiver handling of the "Congestion Window Reduced" (CWR) bit is
also exactly as per [RFC3168] including [RFC3168-ERRATA3639]. That also per [RFC3168] including [RFC3168-ERRATA3639]. That is, on
is, on receipt of a segment with both the CE and CWR bits set, CWR is receipt of a segment with both the CE and CWR bits set, CWR is
processed first and then ECE is processed. processed first and then ECE is processed.
Send immediate Send immediate
ACK with ECE=0 ACK with ECE=0
.----. .-------------. .---. .----. .-------------. .---.
Send 1 ACK / v v | | \ Send 1 ACK / v v | | \
for every | .------. .------. | Send 1 ACK for every | .------. .------. | Send 1 ACK
m packets | | CE=0 | | CE=1 | | for every m packets | | CE=0 | | CE=1 | | for every
with ECE=0 | '------' '------' | m packets with ECE=0 | '------' '------' | m packets
\ | | ^ ^ / with ECE=1 \ | | ^ ^ / with ECE=1
skipping to change at page 6, line 18 skipping to change at page 6, line 33
particular, an observation window ends when all bytes in flight at particular, an observation window ends when all bytes in flight at
the beginning of the window have been acknowledged. the beginning of the window have been acknowledged.
In order to update DCTCP.Alpha, the TCP state variables defined in In order to update DCTCP.Alpha, the TCP state variables defined in
[RFC0793] are used, and three additional TCP state variables are [RFC0793] are used, and three additional TCP state variables are
introduced: introduced:
o DCTCP.WindowEnd: The TCP sequence number threshold for beginning a o DCTCP.WindowEnd: The TCP sequence number threshold for beginning a
new observation window; initialized to SND.UNA. new observation window; initialized to SND.UNA.
o DCTCP.BytesSent: The number of bytes sent during the current o DCTCP.BytesAcked: The number of sent bytes acknowledged during the
observation window; initialized to zero. current observation window; initialized to zero.
o DCTCP.BytesMarked: The number of bytes sent during the current o DCTCP.BytesMarked: The number of bytes sent during the current
observation window that encountered congestion; initialized to observation window that encountered congestion; initialized to
zero. zero.
The congestion estimator on the sender SHOULD process acceptable ACKs The congestion estimator on the sender SHOULD process acceptable ACKs
as follows: as follows:
1. Compute the bytes acknowledged (TCP SACK options [RFC2018] are 1. Compute the bytes acknowledged (TCP SACK options [RFC2018] are
ignored for this computation): ignored for this computation):
BytesAcked = SEG.ACK - SND.UNA BytesAcked = SEG.ACK - SND.UNA
2. Update the bytes sent: 2. Update the bytes sent:
DCTCP.BytesSent += BytesAcked DCTCP.BytesAcked += BytesAcked
3. If the ECE flag is set, update the bytes marked: 3. If the ECE flag is set, update the bytes marked:
DCTCP.BytesMarked += BytesAcked DCTCP.BytesMarked += BytesAcked
4. If the acknowledgment number is less than or equal to 4. If the acknowledgment number is less than or equal to
DCTCP.WindowEnd, stop processing. Otherwise, the end of the DCTCP.WindowEnd, stop processing. Otherwise, the end of the
observation window has been reached, so proceed to update the observation window has been reached, so proceed to update the
congestion estimate as follows: congestion estimate as follows:
5. Compute the congestion level for the current observation window: 5. Compute the congestion level for the current observation window:
M = DCTCP.BytesMarked / DCTCP.BytesSent M = DCTCP.BytesMarked / DCTCP.BytesAcked
6. Update the congestion estimate: 6. Update the congestion estimate:
DCTCP.Alpha = DCTCP.Alpha * (1 - g) + g * M DCTCP.Alpha = DCTCP.Alpha * (1 - g) + g * M
7. Determine the end of the next observation window: 7. Determine the end of the next observation window:
DCTCP.WindowEnd = SND.NXT DCTCP.WindowEnd = SND.NXT
8. Reset the byte counters: 8. Reset the byte counters:
DCTCP.BytesSent = DCTCP.BytesMarked = 0 DCTCP.BytesAcked = DCTCP.BytesMarked = 0
Rather than always halving the congestion window as described in Rather than always halving the congestion window as described in
[RFC3168], when the sender receives an indication of congestion [RFC3168], when the sender receives an indication of congestion
(ECE), the sender SHOULD update cwnd as follows: (ECE), the sender SHOULD update cwnd as follows:
cwnd = cwnd * (1 - DCTCP.Alpha / 2) cwnd = cwnd * (1 - DCTCP.Alpha / 2)
Thus, when no bytes sent experienced congestion, DCTCP.Alpha equals Thus, when no bytes sent experienced congestion, DCTCP.Alpha equals
zero, and cwnd is left unchanged. When all sent bytes experienced zero, and cwnd is left unchanged. When all sent bytes experienced
congestion, DCTCP.Alpha equals one, and cwnd is reduced by half. congestion, DCTCP.Alpha equals one, and cwnd is reduced by half.
skipping to change at page 7, line 39 skipping to change at page 8, line 7
potential misconfigurations. potential misconfigurations.
A DCTCP sender MUST deal with loss episodes in the same way as A DCTCP sender MUST deal with loss episodes in the same way as
conventional TCP. In case of a timeout or fast retransmit or any conventional TCP. In case of a timeout or fast retransmit or any
change in delay (for delay based congestion control), the cwnd and change in delay (for delay based congestion control), the cwnd and
other state variables like ssthresh must be changed in the same way other state variables like ssthresh must be changed in the same way
that a conventional TCP would have changed them. that a conventional TCP would have changed them.
3.4. Handling of SYN, SYN-ACK, RST Packets 3.4. Handling of SYN, SYN-ACK, RST Packets
[RFC3168] requires that a compliant TCP MUST NOT set ECT on SYN or The switching fabric can drop TCP packets that do not have the ECT
SYN-ACK packets. [RFC5562] proposes setting ECT on SYN-ACK packets, set in the IP header. If SYN and SYN-ACK packets for DCTCP
but maintains the restriction of no ECT on SYN packets. Both these connections do not have ECT set, they will be dropped with high
RFCs prohibit ECT in SYN packets due to security concerns regarding probability. For DCTCP connections, the sender SHOULD set ECT for
malicious SYN packets with ECT set. These RFCs, however, are SYN, SYN-ACK and RST packets.
intended for general Internet use, and do not directly apply to a
controlled datacenter environment. The switching fabric can drop TCP
packets that do not have the ECT set in the IP header. If SYN and
SYN-ACK packets for DCTCP connections do not have ECT set, they will
be dropped with high probability. For DCTCP connections, the sender
SHOULD set ECT for SYN, SYN-ACK and RST packets. The security
concerns addressed by both these RFCs might not apply in controlled
environments like datacenters, and it might not be necessary to cater
to both the presence of non-ECN servers.
4. Implementation Issues 4. Implementation Issues
As noted in Section 3.3, the implementation will need to choose a As noted in Section 3.3, the implementation will need to choose a
suitable estimation gain. [DCTCP10] provides a theoretical basis for suitable estimation gain. [DCTCP10] provides a theoretical basis for
selecting the gain. However, it may be more practical to use selecting the gain. However, it may be more practical to use
experimentation to select a suitable gain for a particular network experimentation to select a suitable gain for a particular network
and workload. The Microsoft implementation of DCTCP in Windows and workload. The Microsoft implementation of DCTCP in Windows
Server 2012 uses a fixed estimation gain of 1/16. Server 2012 uses a fixed estimation gain of 1/16.
skipping to change at page 8, line 34 skipping to change at page 8, line 42
ensure a DCTCP sender is always paired with a DCTCP receiver. One ensure a DCTCP sender is always paired with a DCTCP receiver. One
approach is to enable DCTCP based on the IP address of the remote approach is to enable DCTCP based on the IP address of the remote
endpoint. Another approach is to detect connections that transmit endpoint. Another approach is to detect connections that transmit
within the bounds a datacenter. For example, Microsoft Windows within the bounds a datacenter. For example, Microsoft Windows
Server 2012 (and later versions) supports automatic selection of Server 2012 (and later versions) supports automatic selection of
DCTCP if the estimated RTT is less than 10 msec and ECN is DCTCP if the estimated RTT is less than 10 msec and ECN is
successfully negotiated, under the assumption that if the RTT is low, successfully negotiated, under the assumption that if the RTT is low,
then the two endpoints are likely in the same datacenter network. then the two endpoints are likely in the same datacenter network.
[RFC3168] forbids the ECN-marking of pure ACK packets, because of the [RFC3168] forbids the ECN-marking of pure ACK packets, because of the
inability of TCP to mitigate ACK-path congestion and the extra inability of TCP to mitigate ACK-path congestion. RFC 3168 also
advantage to injection attackers that ECN is perceived to offer. For forbids ECN-marking of retransmissions, window probes and RSTs.
the latter reason RFC 3168 also forbids ECN-marking of However, dropping all these control packets - rather than ECN marking
retransmissions, window probes and RSTs. However, dropping all these them - has considerable performance disadvantages. It is RECOMMENDED
control packets - rather than ECN marking them - has considerable that an implementation provide a configuration knob that will cause
performance disadvantages. It is RECOMMENDED that an implementation ECT to be set on such control packes, which can be used in
provide a configuration knob that will cause ECT to be set on such environments where such concerns do not apply.
control packes, which can be used in environments where such concerns
do not apply.
It would be useful to implement DCTCP as additional actions on top of It would be useful to implement DCTCP as additional actions on top of
an existing congestion control algorithm like NewReno. The DCTCP an existing congestion control algorithm like NewReno. The DCTCP
implementation MAY also allow configuration of resetting the value of implementation MAY also allow configuration of resetting the value of
DCTCP.Alpha as part of processing any loss episodes. DCTCP.Alpha as part of processing any loss episodes.
The DCTCP.Alpha calculation as per the formula in Section 3.3 The DCTCP.Alpha calculation as per the formula in Section 3.3
involves fractions. An efficient kernel implementation MAY scale the involves fractions. An efficient kernel implementation MAY scale the
DCTCP.Alpha value for efficient computation using shift operations. DCTCP.Alpha value for efficient computation using shift operations.
For example, if the implementation chooses g as 1/16, multiplications For example, if the implementation chooses g as 1/16, multiplications
skipping to change at page 9, line 18 skipping to change at page 9, line 25
DCTCP.Alpha does not exceed the scaling factor, which would be DCTCP.Alpha does not exceed the scaling factor, which would be
equivalent to greater than 100% congestion. So, DCTCP.Alpha MUST be equivalent to greater than 100% congestion. So, DCTCP.Alpha MUST be
clamped after an update. clamped after an update.
This results in the following computations replacing steps 5 and 6 in This results in the following computations replacing steps 5 and 6 in
Section 3.3, where SCF is the chosen scaling factor (65536 in the Section 3.3, where SCF is the chosen scaling factor (65536 in the
example) and SHF is the shift factor (4 in the example): example) and SHF is the shift factor (4 in the example):
1. Compute the congestion level for the current observation window: 1. Compute the congestion level for the current observation window:
ScaledM = SCF * DCTCP.BytesMarked / DCTCP.BytesSent ScaledM = SCF * DCTCP.BytesMarked / DCTCP.BytesAcked
2. Update the congestion estimate: 2. Update the congestion estimate:
if (DCTCP.Alpha >> SHF) == 0 then DCTCP.Alpha = 0 if (DCTCP.Alpha >> SHF) == 0 then DCTCP.Alpha = 0
DCTCP.Alpha += (ScaledM >> SHF) - (DCTCP.Alpha >> SHF) DCTCP.Alpha += (ScaledM >> SHF) - (DCTCP.Alpha >> SHF)
if DCTCP.Alpha > SCF then DCTCP.Alpha = SCF if DCTCP.Alpha > SCF then DCTCP.Alpha = SCF
5. Deployment Issues 5. Deployment Issues
skipping to change at page 11, line 13 skipping to change at page 11, line 19
negotiation, causing further interoperability issues. negotiation, causing further interoperability issues.
Much like standard TCP, DCTCP is biased against flows with longer Much like standard TCP, DCTCP is biased against flows with longer
RTTs. A method for improving the RTT fairness of DCTCP has been RTTs. A method for improving the RTT fairness of DCTCP has been
proposed in [ADCTCP], but requires additional experimental proposed in [ADCTCP], but requires additional experimental
evaluation. evaluation.
7. Implementation Status 7. Implementation Status
This section documents the implementation status of the specification This section documents the implementation status of the specification
in this document, as recommended by [RFC6982]. in this document, as recommended by [RFC7942].
This document describes DCTCP as implemented in Microsoft Windows This document describes DCTCP as implemented in Microsoft Windows
Server 2012. Since publication of the first versions of this Server 2012. Since publication of the first versions of this
document, the Linux [LINUX] and FreeBSD [FREEBSD] operating systems document, the Linux [LINUX] and FreeBSD [FREEBSD] operating systems
have also implemented support for DCTCP in a way that is believed to have also implemented support for DCTCP in a way that is believed to
follow this document. follow this document.
8. Security Considerations 8. Security Considerations
DCTCP enhances ECN and thus inherits the security considerations DCTCP enhances ECN and thus inherits the security considerations
discussed in [RFC3168]. The processing changes introduced by DCTCP discussed in [RFC3168]. The processing changes introduced by DCTCP
do not exacerbate these considerations or introduce new ones. In do not exacerbate these considerations or introduce new ones. In
particular, with either algorithm, the network infrastructure or the particular, with either algorithm, the network infrastructure or the
remote endpoint can falsely report congestion and thus cause the remote endpoint can falsely report congestion and thus cause the
sender to reduce cwnd. However, this is no worse than what can be sender to reduce cwnd. However, this is no worse than what can be
achieved by simply dropping packets. achieved by simply dropping packets.
[RFC3168] requires that a compliant TCP must not set ECT on SYN or
SYN-ACK packets. [RFC5562] proposes setting ECT on SYN-ACK packets,
but maintains the restriction of no ECT on SYN packets. Both these
RFCs prohibit ECT in SYN packets due to security concerns regarding
malicious SYN packets with ECT set. These RFCs, however, are
intended for general Internet use, and do not directly apply to a
controlled datacenter environment. The security concerns addressed
by both these RFCs might not apply in controlled environments like
datacenters, and it might not be necessary to account for the
presence of non-ECN servers. Since most servers run virtualized in
datacenters, additional security can be imposed in the physical
servers to intercept and drop traffic resembling an attack.
9. IANA Considerations 9. IANA Considerations
This document has no actions for IANA. This document has no actions for IANA.
10. Acknowledgements 10. Acknowledgements
The DCTCP algorithm was originally proposed and analyzed in [DCTCP10] The DCTCP algorithm was originally proposed and analyzed in [DCTCP10]
by Mohammad Alizadeh, Albert Greenberg, Dave Maltz, Jitu Padhye, by Mohammad Alizadeh, Albert Greenberg, Dave Maltz, Jitu Padhye,
Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, and Murari Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, and Murari
Sridharan. Sridharan.
skipping to change at page 12, line 40 skipping to change at page 13, line 13
<http://www.rfc-editor.org/info/rfc5681>. <http://www.rfc-editor.org/info/rfc5681>.
[RFC5562] Kuzmanovic, A., Mondal, A., Floyd, S., and K. [RFC5562] Kuzmanovic, A., Mondal, A., Floyd, S., and K.
Ramakrishnan, "Adding Explicit Congestion Notification Ramakrishnan, "Adding Explicit Congestion Notification
(ECN) Capability to TCP's SYN/ACK Packets", RFC 5562, (ECN) Capability to TCP's SYN/ACK Packets", RFC 5562,
DOI 10.17487/RFC5562, June 2009, DOI 10.17487/RFC5562, June 2009,
<http://www.rfc-editor.org/info/rfc5562>. <http://www.rfc-editor.org/info/rfc5562>.
11.2. Informative References 11.2. Informative References
[RFC6982] Sheffer, Y. and A. Farrel, "Improving Awareness of Running [RFC7942] Sheffer, Y. and A. Farrel, "Improving Awareness of Running
Code: The Implementation Status Section", RFC 6982, Code: The Implementation Status Section", BCP 205,
DOI 10.17487/RFC6982, July 2013, RFC 7942, DOI 10.17487/RFC7942, July 2016,
<http://www.rfc-editor.org/info/rfc6982>. <http://www.rfc-editor.org/info/rfc7942>.
[DCTCP10] Alizadeh, M., Greenberg, A., Maltz, D., Padhye, J., Patel, [DCTCP10] Alizadeh, M., Greenberg, A., Maltz, D., Padhye, J., Patel,
P., Prabhakar, B., Sengupta, S., and M. Sridharan, "Data P., Prabhakar, B., Sengupta, S., and M. Sridharan, "Data
Center TCP (DCTCP)", DOI 10.1145/1851182.1851192, Proc. Center TCP (DCTCP)", DOI 10.1145/1851182.1851192, Proc.
ACM SIGCOMM 2010 Conference (SIGCOMM 10), August 2010, ACM SIGCOMM 2010 Conference (SIGCOMM 10), August 2010,
<http://dl.acm.org/citation.cfm?doid=1851182.1851192>. <http://dl.acm.org/citation.cfm?doid=1851182.1851192>.
[ODCTCP] Kato, M., "Improving Transmission Performance with One- [ODCTCP] Kato, M., "Improving Transmission Performance with One-
Sided Datacenter TCP", M.S. Thesis, Keio University, Sided Datacenter TCP", M.S. Thesis, Keio University,
2014, <http://eggert.org/students/kato-thesis.pdf>. 2014, <http://eggert.org/students/kato-thesis.pdf>.
 End of changes. 26 change blocks. 
64 lines changed or deleted 78 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/