[Docs] [txt|pdf|xml] [Tracker] [WG] [Email] [Diff1] [Diff2] [Nits]

Versions: (draft-raiciu-mptcp-congestion) 00 01 02 03 04 05 06 07 RFC 6356

Internet Engineering Task Force                                C. Raiciu
Internet-Draft                                 University Politehnica of
Intended status: Experimental                                  Bucharest
Expires: January 26, 2012                                     M. Handley
                                                              D. Wischik
                                               University College London
                                                           July 25, 2011


      Coupled Congestion Control for Multipath Transport Protocols
                     draft-ietf-mptcp-congestion-06

Abstract

   Often endpoints are connected by multiple paths, but communications
   are usually restricted to a single path per connection.  Resource
   usage within the network would be more efficient were it possible for
   these multiple paths to be used concurrently.  Multipath TCP is a
   proposal to achieve multipath transport in TCP.

   New congestion control algorithms are needed for multipath transport
   protocols such as Multipath TCP, as single path algorithms have a
   series of issues in the multipath context.  One of the prominent
   problems is that running existing algorithms such as standard TCP
   independently on each path would give the multipath flow more than
   its fair share at a bottleneck link traversed by more than one of its
   subflows.  Further, it is desirable that a source with multiple paths
   available will transfer more traffic using the least congested of the
   paths, achieving a property called resource pooling where a bundle of
   links effectively behaves like one shared link with bigger-capacity.
   This would increase the overall efficiency of the network and also
   its robustness to failure.

   This document presents a congestion control algorithm which couples
   the congestion control algorithms running on different subflows by
   linking their increase functions, and dynamically controls the
   overall aggressiveness of the multipath flow.  The result is a
   practical algorithm that is fair to TCP at bottlenecks while moving
   traffic away from congested links.

Status of this Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-



Raiciu, et al.          Expires January 19, 2012                [Page 1]

Internet-Draft          MPTCP Congestion Control               July 2011


   Drafts is at http://datatracker.ietf.org/drafts/current/.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   This Internet-Draft will expire on January 19, 2012.

Copyright Notice

   Copyright (c) 2011 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.




























Raiciu, et al.          Expires January 19, 2012                [Page 2]

Internet-Draft          MPTCP Congestion Control               July 2011


Table of Contents

   1.  Requirements Language  . . . . . . . . . . . . . . . . . . . .  4
   2.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  4
   3.  Coupled Congestion Control Algorithm . . . . . . . . . . . . .  6
   4.  Implementation Considerations  . . . . . . . . . . . . . . . .  7
     4.1.  Computing alpha in Practice  . . . . . . . . . . . . . . .  8
     4.2.  Implementation Considerations when CWND is Expressed
           in Packets . . . . . . . . . . . . . . . . . . . . . . . .  9
   5.  Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 10
   6.  Security Considerations  . . . . . . . . . . . . . . . . . . . 11
   7.  Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 11
   8.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . . 11
   9.  References . . . . . . . . . . . . . . . . . . . . . . . . . . 11
     9.1.  Normative References . . . . . . . . . . . . . . . . . . . 11
     9.2.  Informative References . . . . . . . . . . . . . . . . . . 11
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 12


































Raiciu, et al.          Expires January 19, 2012                [Page 3]

Internet-Draft          MPTCP Congestion Control               July 2011


1.  Requirements Language

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in RFC 2119 [RFC2119] .


2.  Introduction

   Multipath TCP (MPTCP, [I-D.ford-mptcp-multiaddressed]) is a set of
   extensions to regular TCP [RFC0793] that allows one TCP connection to
   be spread across multiple paths.  MPTCP distributes load through the
   creation of separate "subflows" across potentially disjoint paths.

   How should congestion control be performed for multipath TCP?  First,
   each subflow must have its own congestion control state (i.e. cwnd)
   so that capacity on that path is matched by offered load.  The
   simplest way to achieve this goal is to simply run standard TCP
   congestion control on each subflow.  However this solution is
   unsatisfactory as it gives the multipath flow an unfair share when
   the paths taken by its different subflows share a common bottleneck.

   Bottleneck fairness is just one requirement multipath congestion
   control should meet.  The following three goals capture the desirable
   properties of a practical multipath congestion control algorithm:

   o  Goal 1 (Improve Throughput) A multipath flow should perform at
      least as well as a single path flow would on the best of the paths
      available to it.

   o  Goal 2 (Do no harm) A multipath flow should not take up more
      capacity from any of the resources shared by its different paths,
      than if it was a single flow using only one of these paths.  This
      guarantees it will not unduly harm other flows.

   o  Goal 3 (Balance congestion) A multipath flow should move as much
      traffic as possible off its most congested paths, subject to
      meeting the first two goals.

   Goals 1 and 2 together ensure fairness at the bottleneck.  Goal 3
   captures the concept of resource pooling [WISCHIK]: if each multipath
   flow sends more data through its least congested path, the traffic in
   the network will move away from congested areas.  This improves
   robustness and overall throughput, among other things.  The way to
   achieve resource pooling is to effectively "couple" the congestion
   control loops for the different subflows.

   We propose an algorithm that couples the additive increase function



Raiciu, et al.          Expires January 19, 2012                [Page 4]

Internet-Draft          MPTCP Congestion Control               July 2011


   of the subflows, and uses unmodified TCP behavior in case of a drop.
   The algorithm relies on the traditional TCP mechanisms to detect
   drops, to retransmit data, etc.

   Detecting shared bottlenecks reliably is quite difficult, but is just
   one part of a bigger question.  This bigger question is how much
   bandwidth a multipath user should use in total, even if there is no
   shared bottleneck.

   The congestion controller aims to set the multipath flow's aggregate
   bandwidth to be the same as a regular TCP flow would get on the best
   path available to the multipath flow.  To estimate the bandwidth of a
   regular TCP flow, the multipath flow estimates loss rates and round
   trip times and computes the target rate.  Then it adjusts the overall
   aggresiveness (parameter alpha) to achieve the desired rate.

   While the mechanism above applies always, its effect depends on
   whether the multipath TCP flow influences or does not influence the
   link loss rates (low vs. high statistical multiplexing).  If MPTCP
   does not influence link loss rates, MPTCP will get the same
   throughput as TCP on the best path.  In cases with low statistical
   multiplexing, where the multipath flow influences the loss rates on
   the path, the multipath throughput will be strictly higher than a
   single TCP would get on any of the paths.  In particular, if using
   two idle paths, multipath throughput will be sum of the two paths'
   throughput.

   This algorithm ensures bottleneck fairness and fairness in the
   broader, network sense.  We acknowledge that current TCP fairness
   criteria are far from ideal, but a multipath TCP needs to be
   deployable in the current Internet.  If needed, new fairness criteria
   can be implemented by the same algorithm we propose by appropriately
   scaling the overall aggressiveness.

   It is intended that the algorithm presented here can be applied to
   other multipath transport protocols, such as alternative multipath
   extensions to TCP, or indeed any other congestion-aware transport
   protocols.  However, for the purposes of example this document will,
   where appropriate, refer to the MPTCP protocol.

   The design decisions and evaluation of the congestion control
   algorithm are published in [NSDI].

   The algorithm presented here only extends standard TCP congestion
   control for multipath operation.  It is foreseeable that other
   congestion controllers will be implemented for multipath transport to
   achieve the bandwidth-scaling properties of the newer congestion
   control algorithms for regular TCP (such as Compound TCP and Cubic).



Raiciu, et al.          Expires January 19, 2012                [Page 5]

Internet-Draft          MPTCP Congestion Control               July 2011


3.  Coupled Congestion Control Algorithm

   The algorithm we present only applies to the increase phase of the
   congestion avoidance state specifying how the window inflates upon
   receiving an ack.  The slow start, fast retransmit, and fast recovery
   algorithms, as well as the multiplicative decrease of the congestion
   avoidance state are the same as in standard TCP[RFC5681].

   Let cwnd_i be the congestion window on the subflow i.  Let tot_cwnd
   be the sum of the congestion windows of all subflows in the
   connection.  Let p_i, rtt_i and mss_i be the loss rate, round trip
   time (i.e. smoothed round trip time estimate used by TCP) and maximum
   segment size on subflow i.

   We assume throughout this document that the congestion window is
   maintained in bytes, unless otherwise specified.  We briefly describe
   the algorithm for packet-based implementations of cwnd in section
   Section 4.2.

   Our proposed "Linked Increases" algorithm MUST:

   o  For each ack received on subflow i, increase cwnd_i by min
      (alpha*bytes_acked*mss_i/tot_cwnd , bytes_acked*mss_i/cwnd_i )

   The increase formula takes the minimum between the computed increase
   for the multipath subflow (first argument to min), and the increase
   TCP would get in the same scenario (the second argument).  In this
   way, we ensure that any multipath subflow cannot be more aggressive
   than a TCP flow in the same circumstances, hence achieving goal 2 (do
   no harm).

   "alpha" is a parameter of the algorithm that describes the
   aggresiveness of the multipath flow.  To meet Goal 1 (improve
   throughput), the value of alpha is chosen such that the aggregate
   throughput of the multipath flow is equal to the throughput a TCP
   flow would get if it ran on the best path.

   To get an intuition of what the algorithm is trying to do, let's take
   the case where all the subflows have the same round trip time and
   MSS.  In this case the algorithm will grow the total window by
   approximately alpha*MSS per RTT.  This increase is distributed to the
   individual flows according to their instantaneous window size.
   Subflow i will increase by alpha*cwnd_i/tot_cwnd segments per RTT.

   Note that, as in standard TCP, when tot_cwnd is large the increase
   may be 0.  In this case the increase MUST be set to 1.  We discuss
   how to implement this formula in practice in the next section.




Raiciu, et al.          Expires January 19, 2012                [Page 6]

Internet-Draft          MPTCP Congestion Control               July 2011


   We assume implementations use an approach similar to appropriate byte
   counting (ABC, [RFC3465]), where the bytes_acked variable records the
   number of bytes newly acknowledged.  If this is not true, bytes_acked
   SHOULD be set to mss_i.

   To compute tot_cwnd, it is an easy mistake to sum up cwnd_i across
   all subflows: when a flow is in fast retransmit, its cwnd is
   typically inflated and no longer represents the real congestion
   window.  The correct behavior is to use the ssthresh value for flows
   in fast retransmit when computing tot_cwnd.  To cater for connections
   that are app limited, the computation should consider the minimum
   between flight_size_i and cwnd_i, and flight_size_i and ssthresh_i
   where appropriate.

   The total throughput of a multipath flow depends on the value of
   alpha and the loss rates, maximum segment sizes and round trip times
   of its paths.  Since we require that the total throughput is no worse
   than the throughput a single TCP would get on the best path, it is
   impossible to choose a-priori a single value of alpha that achieves
   the desired throughput in every occasion.  Hence, alpha must be
   computed based on the observed properties of the paths.

   The formula to compute alpha is:

                                      cwnd_i
                                 max --------
                                  i         2
                                      rtt_i
             alpha = tot_cwnd * ----------------
                               /      cwnd_i \ 2
                               | sum ---------|
                               \  i   rtt_i  /


   The formula is derived by equalizing the rate of the multipath flow
   with the rate of a TCP running on the best path, and solving for
   alpha.


4.  Implementation Considerations

   The formula for alpha above implies that alpha is a floating point
   value.  This would require performing costly floating point
   operations whenever an ACK is received, Further, in many kernels
   floating point operations are disabled.  There is an easy way to
   approximate the above calculations using integer arithmetic.





Raiciu, et al.          Expires January 19, 2012                [Page 7]

Internet-Draft          MPTCP Congestion Control               July 2011


4.1.  Computing alpha in Practice

   Let alpha_scale be an integer.  When computing alpha, use alpha_scale
   * tot_cwnd instead of tot_cwnd, and do all the operations in integer
   arithmetic.

   Then, scale down the increase per ack by alpha_scale.  The algorithm
   is:

   o  For each ack received on subflow i, increase cwnd_i by min (
      (alpha*bytes_acked*mss_i/tot_cwnd)/alpha_scale ,
      bytes_acked*mss_i/cwnd_i )

   Alpha scale denotes the precision we want for computing alpha.
   Observe that the errors in computing the numerator or the denominator
   in the formula for alpha are quite small, as the cwnd in bytes is
   typically much larger than the RTT (measured in ms).

   With these changes, all the operations can be done using integer
   arithmetic.  We propose alpha_scale be a small power of two, to allow
   using faster shift operations instead of multiplication and division.
   Our experiments show that using alpha_scale=512 works well in a wide
   range of scenarios.  Increasing alpha_scale increases precision, but
   also increases the risk of overflow when computing alpha.  Using
   64bit operations would solve this issue.  Another option is to
   dynamically adjust alpha_scale when computing alpha; in this way we
   avoid overflow and obtain maximum precision.

   It is possible to implement the algorithm by calculating tot_cwnd on
   each ack, however this would be costly especially when the number of
   subflows is large.  To avoid this overhead the implementation MAY
   choose to maintain a new per connection state variable called
   tot_cwnd.  If it does so, the implementation will update tot_cwnd
   value whenever the individual subflows' windows are updated.
   Updating only requires one more addition or subtraction operation
   compared to the regular, per subflow congestion control code, so its
   performance impact should be minimal.

   Computing alpha per ack is also costly.  We propose alpha to be a per
   connection variable, computed whenever there is a drop and once per
   RTT otherwise.  More specifically, let cwnd_new be the new value of
   the congestion window after it is inflated or after a drop.  Update
   alpha only if cwnd_i/mss_i != cwnd_new_i/mss_i.

   In certain cases with small RTTs, computing alpha can still be
   expensive.  We observe that if RTTs were constant, it is sufficient
   to compute alpha once per drop, as alpha does not change between
   drops (the insight here is that cwnd_i/cwnd_j = constant as long as



Raiciu, et al.          Expires January 19, 2012                [Page 8]

Internet-Draft          MPTCP Congestion Control               July 2011


   both windows increase).  Experimental results show that even if round
   trip times are not constant, using average round trip time per
   sawtooth instead of instantaneous round trip time (i.e.  TCP's
   smoothed RTT estimator) gives good precision for computing alpha.
   Hence, it is possible to compute alpha only once per drop according
   to the formula above, by replacing rtt_i with rtt_avg_i.

   If using average round trip time, rtt_avg_i will be computed by
   sampling the rtt_i whenever the window can accommodate one more
   packet, i.e. when cwnd / mss < (cwnd+increase)/mss.  The samples are
   averaged once per sawtooth into rtt_avg_i.  This sampling ensures
   that there is no sampling bias for larger windows.

   Given tot_cwnd and alpha, the congestion control algorithm is run for
   each subflow independently, with similar complexity to the standard
   TCP increase code [RFC5681].

4.2.  Implementation Considerations when CWND is Expressed in Packets

   When the congestion control algorithm maintains cwnd in packets
   rather than bytes, the algorithms above must change to take into
   account path mss.

   To compute the increase when an ack is received, the implementation
   for multipath congestion control is a simple extension of the
   standard TCP code.  In standard TCP cwnd_cnt is an additional state
   variable that tracks the number of segments acked since the last cwnd
   increment; cwnd is incremented only when cwnd_cnt > cwnd; then
   cwnd_cnt is set to 0.

   In the multipath case, cwnd_cnt_i is maintained for each subflow as
   above, and cwnd_i is increased by 1 when cwnd_cnt_i > max(alpha_scale
   * tot_cwnd / alpha, cwnd_i).

   When computing alpha for packet-based stacks, the errors in computing
   the terms in the denominator are larger (this is because cwnd is much
   smaller and rtt may be comparatively large).  Let max be the index of
   the subflow used in the numerator.  To reduce errors, it is easiest
   to move rtt_max (once calculated) from the numerator to the
   denominator, obtaining the equivalent formula below.

                                               cwnd_max
        alpha = alpha_scale * tot_cwnd * -----------------------
                                        /     rtt_max * cwnd_i \ 2
                                        | sum -----------------|
                                        \  i        rtt_i      /





Raiciu, et al.          Expires January 19, 2012                [Page 9]

Internet-Draft          MPTCP Congestion Control               July 2011


   Note that the formula for computing alpha does not take into account
   path mss, and is the same for stacks that keep cwnd in bytes or
   packets.  With this formula, the algorithm for computing alpha will
   match the rate of TCP on the best path in B/s for byte-oriented
   stacks, and in packets/s in packet-based stacks.  In practice, mss
   rarely changes between paths so this shouldn't be a problem.

   However, it is simple to derive formulae allowing packet-based stacks
   to achieve byte rate fairness (and viceversa) if needed.  In
   particular, for packet-based stacks wanting byte-rate fairness, the
   formula above changes as follows: cwnd_max is replaced by cwnd_max *
   mss_max * mss_max, while cwnd_i is replaced with cwnd_i * mss_i.


5.  Discussion

   The algorithm we've presented fully achieves Goals 1 and 2, but does
   not achieve full resource pooling (Goal 3).  Resource pooling
   requires that no traffic should be transferred on links with higher
   loss rates.  To achieve perfect resource pooling, one must couple
   both increase and decrease of congestion windows across subflows, as
   in [KELLY].

   There are a few problems with such a fully-coupled controller.
   First, it will probe insufficiently paths with high loss rates, and
   will fail to detect free capacity when it becomes available.  Second,
   such controllers tend to exhibit "flappiness": when the paths have
   similar levels of congestion, the congestion controller will tend to
   allocate all the window to one random subflow, and allocate zero
   window to the other subflows.  The controller will perform random
   flips between these stable points.  This doesn't seem desirable in
   general, and is particularly bad when the achieved rates depend on
   the RTT (as in the current Internet): in such a case, the resulting
   rate with fluctuate unpredictably depending on which state the
   controller is in, hence violating Goal 1.

   By only coupling increases our proposal probes high-loss paths,
   detecting free capacity quicker.  Our proposal does not suffer from
   flappiness but also achieves less resource pooling.  The algorithm
   will allocate window to the subflows such that p_i * cwnd_i =
   constant, for all i.  Thus, when the loss rates of the subflows are
   equal, each subflow will get an equal window, removing flappiness.
   When the loss rates differ, progressively more window will be
   allocated to the flow with the lower loss rate.  In contrast, perfect
   resource pooling requires that all the window should be allocated on
   the path with the lowest loss rate.  Further details can be found in
   [NSDI].




Raiciu, et al.          Expires January 19, 2012               [Page 10]

Internet-Draft          MPTCP Congestion Control               July 2011


6.  Security Considerations

   None.

   Detailed security analysis for the Multipath TCP protocol itself is
   included in [I-D.ford-mptcp-multiaddressed] and RFC 6181 [RFC6181]


7.  Acknowledgements

   We thank Christoph Paasch for his suggestions for computing alpha in
   packet-based stacks.  The authors are supported by Trilogy
   (http://www.trilogy-project.org), a research project (ICT-216372)
   partially funded by the European Community under its Seventh
   Framework Program.  The views expressed here are those of the
   author(s) only.  The European Commission is not liable for any use
   that may be made of the information in this document.


8.  IANA Considerations

   This document does not require any action from IANA.


9.  References

9.1.  Normative References

   [RFC0793]  Postel, J., "Transmission Control Protocol", STD 7,
              RFC 793, September 1981.

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119, March 1997.

   [RFC5681]  Allman, M., Paxson, V., and E. Blanton, "TCP Congestion
              Control", RFC 5681, September 2009.

9.2.  Informative References

   [I-D.ford-mptcp-multiaddressed]
              Ford, A., Raiciu, C., and M. Handley, "TCP Extensions for
              Multipath Operation with Multiple Addresses",
              draft-ford-mptcp-multiaddressed-04 (work in progress),
              March 2010.

   [KELLY]    Kelly, F. and T. Voice, "Stability of end-to-end
              algorithms for joint routing and rate control", ACM
              SIGCOMM CCR vol. 35 num. 2, pp. 5-12, 2005,



Raiciu, et al.          Expires January 19, 2012               [Page 11]

Internet-Draft          MPTCP Congestion Control               July 2011


              <http://portal.acm.org/citation.cfm?id=1064415>.

   [NSDI]     Wischik, D., Raiciu, C., Greenhalgh, A., and M. Handley,
              "Design, Implementation and Evaluation of Congestion
              Control for Multipath TCP", Usenix NSDI , March 2011, <htt
              p://www.cs.ucl.ac.uk/staff/c.raiciu/files/mptcp-nsdi.pdf>.

   [RFC3465]  Allman, M., "TCP Congestion Control with Appropriate Byte
              Counting (ABC)", RFC 3465, February 2003.

   [RFC6181]  Bagnulo, M., "Threat Analysis for TCP Extensions for
              Multipath Operation with Multiple Addresses", RFC 6181,
              March 2011.

   [WISCHIK]  Wischik, D., Handley, M., and M. Bagnulo Braun, "The
              Resource Pooling Principle", ACM SIGCOMM CCR vol. 38 num.
              5, pp. 47-52, October 2008,
              <http://ccr.sigcomm.org/online/files/p47-handleyA4.pdf>.


Authors' Addresses

   Costin Raiciu
   University Politehnica of Bucharest
   Splaiul Independentei 313
   Bucharest
   Romania

   Email: costin.raiciu@cs.pub.ro


   Mark Handley
   University College London
   Gower Street
   London  WC1E 6BT
   UK

   Email: m.handley@cs.ucl.ac.uk


   Damon Wischik
   University College London
   Gower Street
   London  WC1E 6BT
   UK

   Email: d.wischik@cs.ucl.ac.uk




Raiciu, et al.          Expires January 19, 2012               [Page 12]


Html markup produced by rfcmarkup 1.108, available from http://tools.ietf.org/tools/rfcmarkup/