One document matched: draft-floyd-tsvwg-cc-alt-00.ps


%!PS-Adobe-3.0
%%BoundingBox: 24 24 588 768
%%Title: Enscript Output
%%For: Sally Floyd
%%Creator: GNU enscript 1.6.1
%%CreationDate: Tue Dec  5 21:04:28 2006
%%Orientation: Portrait
%%Pages: (atend)
%%DocumentMedia: Letter 612 792 0 () ()
%%DocumentNeededResources: (atend)
%%EndComments
%%BeginProlog
%%BeginResource: procset Enscript-Prolog 1.6 1
%
% Procedures.
%

/_S {	% save current state
  /_s save def
} def
/_R {	% restore from saved state
  _s restore
} def

/S {	% showpage protecting gstate
  gsave
  showpage
  grestore
} bind def

/MF {	% fontname newfontname -> -	make a new encoded font
  /newfontname exch def
  /fontname exch def

  /fontdict fontname findfont def
  /newfont fontdict maxlength dict def

  fontdict {
    exch
    dup /FID eq {
      % skip FID pair
      pop pop
    } {
      % copy to the new font dictionary
      exch newfont 3 1 roll put
    } ifelse
  } forall

  newfont /FontName newfontname put

  % insert only valid encoding vectors
  encoding_vector length 256 eq {
    newfont /Encoding encoding_vector put
  } if

  newfontname newfont definefont pop
} def

/SF { % fontname width height -> -	set a new font
  /height exch def
  /width exch def

  findfont
  [width 0 0 height 0 0] makefont setfont
} def

/SUF { % fontname width height -> -	set a new user font
  /height exch def
  /width exch def

  /F-gs-user-font MF
  /F-gs-user-font width height SF
} def

/M {moveto} bind def
/s {show} bind def

/Box {	% x y w h -> -			define box path
  /d_h exch def /d_w exch def /d_y exch def /d_x exch def
  d_x d_y  moveto
  d_w 0 rlineto
  0 d_h rlineto
  d_w neg 0 rlineto
  closepath
} def

/bgs {	% x y height blskip gray str -> -	show string with bg color
  /str exch def
  /gray exch def
  /blskip exch def
  /height exch def
  /y exch def
  /x exch def

  gsave
    x y blskip sub str stringwidth pop height Box
    gray setgray
    fill
  grestore
  x y M str s
} def

% Highlight bars.
/highlight_bars {	% nlines lineheight output_y_margin gray -> -
  gsave
    setgray
    /ymarg exch def
    /lineheight exch def
    /nlines exch def

    % This 2 is just a magic number to sync highlight lines to text.
    0 d_header_y ymarg sub 2 sub translate

    /cw d_output_w cols div def
    /nrows d_output_h ymarg 2 mul sub lineheight div cvi def

    % for each column
    0 1 cols 1 sub {
      cw mul /xp exch def

      % for each rows
      0 1 nrows 1 sub {
        /rn exch def
        rn lineheight mul neg /yp exch def
        rn nlines idiv 2 mod 0 eq {
	  % Draw highlight bar.  4 is just a magic indentation.
	  xp 4 add yp cw 8 sub lineheight neg Box fill
	} if
      } for
    } for

  grestore
} def

% Line highlight bar.
/line_highlight {	% x y width height gray -> -
  gsave
    /gray exch def
    Box gray setgray fill
  grestore
} def

% Column separator lines.
/column_lines {
  gsave
    .1 setlinewidth
    0 d_footer_h translate
    /cw d_output_w cols div def
    1 1 cols 1 sub {
      cw mul 0 moveto
      0 d_output_h rlineto stroke
    } for
  grestore
} def

% Column borders.
/column_borders {
  gsave
    .1 setlinewidth
    0 d_footer_h moveto
    0 d_output_h rlineto
    d_output_w 0 rlineto
    0 d_output_h neg rlineto
    closepath stroke
  grestore
} def

% Do the actual underlay drawing
/draw_underlay {
  ul_style 0 eq {
    ul_str true charpath stroke
  } {
    ul_str show
  } ifelse
} def

% Underlay
/underlay {	% - -> -
  gsave
    0 d_page_h translate
    d_page_h neg d_page_w atan rotate

    ul_gray setgray
    ul_font setfont
    /dw d_page_h dup mul d_page_w dup mul add sqrt def
    ul_str stringwidth pop dw exch sub 2 div ul_h_ptsize -2 div moveto
    draw_underlay
  grestore
} def

/user_underlay {	% - -> -
  gsave
    ul_x ul_y translate
    ul_angle rotate
    ul_gray setgray
    ul_font setfont
    0 0 ul_h_ptsize 2 div sub moveto
    draw_underlay
  grestore
} def

% Page prefeed
/page_prefeed {		% bool -> -
  statusdict /prefeed known {
    statusdict exch /prefeed exch put
  } {
    pop
  } ifelse
} def

% Wrapped line markers
/wrapped_line_mark {	% x y charwith charheight type -> -
  /type exch def
  /h exch def
  /w exch def
  /y exch def
  /x exch def

  type 2 eq {
    % Black boxes (like TeX does)
    gsave
      0 setlinewidth
      x w 4 div add y M
      0 h rlineto w 2 div 0 rlineto 0 h neg rlineto
      closepath fill
    grestore
  } {
    type 3 eq {
      % Small arrows
      gsave
        .2 setlinewidth
        x w 2 div add y h 2 div add M
        w 4 div 0 rlineto
        x w 4 div add y lineto stroke

        x w 4 div add w 8 div add y h 4 div add M
        x w 4 div add y lineto
	w 4 div h 8 div rlineto stroke
      grestore
    } {
      % do nothing
    } ifelse
  } ifelse
} def

% EPSF import.

/BeginEPSF {
  /b4_Inc_state save def    		% Save state for cleanup
  /dict_count countdictstack def	% Count objects on dict stack
  /op_count count 1 sub def		% Count objects on operand stack
  userdict begin
  /showpage { } def
  0 setgray 0 setlinecap
  1 setlinewidth 0 setlinejoin
  10 setmiterlimit [ ] 0 setdash newpath
  /languagelevel where {
    pop languagelevel
    1 ne {
      false setstrokeadjust false setoverprint
    } if
  } if
} bind def

/EndEPSF {
  count op_count sub { pos } repeat	% Clean up stacks
  countdictstack dict_count sub { end } repeat
  b4_Inc_state restore
} bind def

% Check PostScript language level.
/languagelevel where {
  pop /gs_languagelevel languagelevel def
} {
  /gs_languagelevel 1 def
} ifelse
%%EndResource
%%BeginResource: procset Enscript-Encoding-88591 1.6 1
/encoding_vector [
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/space        	/exclam       	/quotedbl     	/numbersign   	
/dollar       	/percent      	/ampersand    	/quoteright   	
/parenleft    	/parenright   	/asterisk     	/plus         	
/comma        	/hyphen       	/period       	/slash        	
/zero         	/one          	/two          	/three        	
/four         	/five         	/six          	/seven        	
/eight        	/nine         	/colon        	/semicolon    	
/less         	/equal        	/greater      	/question     	
/at           	/A            	/B            	/C            	
/D            	/E            	/F            	/G            	
/H            	/I            	/J            	/K            	
/L            	/M            	/N            	/O            	
/P            	/Q            	/R            	/S            	
/T            	/U            	/V            	/W            	
/X            	/Y            	/Z            	/bracketleft  	
/backslash    	/bracketright 	/asciicircum  	/underscore   	
/quoteleft    	/a            	/b            	/c            	
/d            	/e            	/f            	/g            	
/h            	/i            	/j            	/k            	
/l            	/m            	/n            	/o            	
/p            	/q            	/r            	/s            	
/t            	/u            	/v            	/w            	
/x            	/y            	/z            	/braceleft    	
/bar          	/braceright   	/tilde        	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/.notdef      	/.notdef      	/.notdef      	/.notdef      	
/space        	/exclamdown   	/cent         	/sterling     	
/currency     	/yen          	/brokenbar    	/section      	
/dieresis     	/copyright    	/ordfeminine  	/guillemotleft	
/logicalnot   	/hyphen       	/registered   	/macron       	
/degree       	/plusminus    	/twosuperior  	/threesuperior	
/acute        	/mu           	/paragraph    	/bullet       	
/cedilla      	/onesuperior  	/ordmasculine 	/guillemotright	
/onequarter   	/onehalf      	/threequarters	/questiondown 	
/Agrave       	/Aacute       	/Acircumflex  	/Atilde       	
/Adieresis    	/Aring        	/AE           	/Ccedilla     	
/Egrave       	/Eacute       	/Ecircumflex  	/Edieresis    	
/Igrave       	/Iacute       	/Icircumflex  	/Idieresis    	
/Eth          	/Ntilde       	/Ograve       	/Oacute       	
/Ocircumflex  	/Otilde       	/Odieresis    	/multiply     	
/Oslash       	/Ugrave       	/Uacute       	/Ucircumflex  	
/Udieresis    	/Yacute       	/Thorn        	/germandbls   	
/agrave       	/aacute       	/acircumflex  	/atilde       	
/adieresis    	/aring        	/ae           	/ccedilla     	
/egrave       	/eacute       	/ecircumflex  	/edieresis    	
/igrave       	/iacute       	/icircumflex  	/idieresis    	
/eth          	/ntilde       	/ograve       	/oacute       	
/ocircumflex  	/otilde       	/odieresis    	/divide       	
/oslash       	/ugrave       	/uacute       	/ucircumflex  	
/udieresis    	/yacute       	/thorn        	/ydieresis    	
] def
%%EndResource
%%EndProlog
%%BeginSetup
%%IncludeResource: font Courier-Bold
%%IncludeResource: font Courier
/HFpt_w 10 def
/HFpt_h 10 def
/Courier-Bold /HF-gs-font MF
/HF /HF-gs-font findfont [HFpt_w 0 0 HFpt_h 0 0] makefont def
/Courier /F-gs-font MF
/F-gs-font 10 10 SF
/#copies 1 def
% Pagedevice definitions:
gs_languagelevel 1 gt {
  <<
    /PageSize [612 792] 
  >> setpagedevice
} if
/d_page_w 564 def
/d_page_h 744 def
/d_header_x 0 def
/d_header_y 744 def
/d_header_w 564 def
/d_header_h 0 def
/d_footer_x 0 def
/d_footer_y 0 def
/d_footer_w 564 def
/d_footer_h 0 def
/d_output_w 564 def
/d_output_h 744 def
/cols 1 def
%%EndSetup
%%Page: (1) 1
%%BeginPageSetup
_S
24 24 translate
/pagenum 1 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
5 687 M
(Network Working Group                                           S. Floyd) s
5 676 M
(Internet-Draft                                                 M. Allman) s
5 665 M
(Expires: June 2006                                           ICIR / ICSI) s
5 654 M
(                                                            December 2006) s
5 621 M
(               Specifying New Congestion Control Algorithms) s
5 610 M
(                     draft-floyd-tsvwg-cc-alt-00.txt) s
5 588 M
(Status of this Memo) s
5 566 M
(     By submitting this Internet-Draft, each author represents that any) s
5 555 M
(     applicable patent or other IPR claims of which he or she is aware) s
5 544 M
(     have been or will be disclosed, and any of which he or she becomes) s
5 533 M
(     aware will be disclosed, in accordance with Section 6 of BCP 79.) s
5 511 M
(     Internet-Drafts are working documents of the Internet Engineering) s
5 500 M
(     Task Force \(IETF\), its areas, and its working groups.  Note that) s
5 489 M
(     other groups may also distribute working documents as) s
5 478 M
(     Internet-Drafts.) s
5 456 M
(     Internet-Drafts are draft documents valid for a maximum of six) s
5 445 M
(     months and may be updated, replaced, or obsoleted by other documents) s
5 434 M
(     at any time.  It is inappropriate to use Internet-Drafts as) s
5 423 M
(     reference material or to cite them other than as "work in progress.") s
5 401 M
(     The list of current Internet-Drafts can be accessed at) s
5 390 M
(     http://www.ietf.org/ietf/1id-abstracts.txt.) s
5 368 M
(     The list of Internet-Draft Shadow Directories can be accessed at) s
5 357 M
(     http://www.ietf.org/shadow.html.) s
5 335 M
(Copyright Notice) s
5 313 M
(     Copyright \(C\) The Internet Society \(2006\).) s
5 291 M
(Abstract) s
5 269 M
(     The IETF's standard congestion control schemes have been widely) s
5 258 M
(     shown to be inadequate for various environments \(e.g., high-speed or) s
5 247 M
(     wireless networks\).  Recent research has yielded many alternate) s
5 236 M
(     congestion control schemes.  Using these new congestion control) s
5 225 M
(     schemes in the global Internet has possible ramifications to both) s
5 214 M
(     the network and to traffic using the currently standardized) s
5 203 M
(     congestion control.  Therefore, the IETF must proceed with caution) s
5 192 M
(     when dealing with alternate congestion control proposals.  The goal) s
5 181 M
(     of this document is to provide guidance for considering alternate) s
5 170 M
(     congestion control algorithms within the IETF.) s
5 148 M
(     TO BE DELETED BY THE RFC EDITOR UPON PUBLICATION:) s
5 126 M
(     Changes from draft-floyd-cc-alt-00.txt:) s
5 104 M
(     * Changed the name to draft-floyd-tsvwg-cc-alt-00.txt.) s
5 71 M
(Expires: June 2006                                              [Page 1]) s
_R
S
%%Page: (2) 2
%%BeginPageSetup
_S
24 24 translate
/pagenum 2 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
5 720 M
(draft-floyd-tsvwg-cc-alt-00.txt                            December 2006) s
5 698 M
(     * Added a bullet about incremental deployment.  Feedback from) s
5 687 M
(       Colin Perkins) s
5 665 M
(     * Clarified the fairness section;  this section is not saying) s
5 654 M
(       that strict TCP-friendliness is a requirement.) s
5 632 M
(     * Clarified that as an alternative to Full Backoff, a flow) s
5 621 M
(       could stop sending when the packet drop rate is above a) s
5 610 M
(       certain threshold.) s
5 588 M
(     * Clarified that the Full Backoff bullet does not require) s
5 577 M
(       that different flows with different round-trip times) s
5 566 M
(       use the same criteria about when they should back off) s
5 555 M
(       to one packet per round-trip time or less.) s
5 533 M
(     * Added a paragraph about Informational RFCs.) s
5 511 M
(     * Added a bullet about response to transient events, including) s
5 500 M
(       routing events or moving from a private to a shared network.) s
5 478 M
(     END OF NOTES TO BE DELETED.) s
5 445 M
(1.  Introduction) s
5 423 M
(     This document provides guidelines for the IETF to use when) s
5 412 M
(     evaluating suggested congestion control algorithms that) s
5 401 M
(     significantly differ from the general congestion control principles) s
5 390 M
(     outlined in [RFC2914].  The guidance is intended to be useful to) s
5 379 M
(     authors proposing alternate congestion control and for the IETF) s
5 368 M
(     community when evaluating whether a proposal is appropriate for) s
5 357 M
(     publication in the RFC series.) s
5 335 M
(     This document does not give hard-and-fast rules for what makes for) s
5 324 M
(     an appropriate congestion control scheme.  Rather, the document) s
5 313 M
(     provides a set of criteria that should be considered and weighed by) s
5 302 M
(     the IETF in the context of each proposal.  The high-order criteria) s
5 291 M
(     for any new proposal is that a serious scientific study of the pros) s
5 280 M
(     and cons of the proposal needs to have been done such that the IETF) s
5 269 M
(     has a well rounded set of information to consider.) s
5 247 M
(     After initial studies, we encourage authors to write a specification) s
5 236 M
(     of their proposals for publication in the RFC series to allow others) s
5 225 M
(     to concretely understand and investigate the wealth of proposals in) s
5 214 M
(     this space.) s
5 192 M
(2.  Status) s
5 170 M
(     Following the lead of HighSpeed TCP, alternate congestion control) s
5 159 M
(     algorithms are expected to be published as "Experimental" RFCs until) s
5 148 M
(     such time that the community better understands the solution space.) s
5 137 M
(     Traditionally, the meaning of "Experimental" status has varied in) s
5 126 M
(     its use and interpretation.  As part of this document we define two) s
5 115 M
(     classes of congestion control proposals that can be published with) s
5 93 M
(Expires: June 2006                                              [Page 2]) s
_R
S
%%Page: (3) 3
%%BeginPageSetup
_S
24 24 translate
/pagenum 3 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
5 720 M
(draft-floyd-tsvwg-cc-alt-00.txt                            December 2006) s
5 698 M
(     the "Experimental" status.  The first class is algorithms that are) s
5 687 M
(     judged to be safe to deploy in the global Internet and further) s
5 676 M
(     investigated in that environment.  The second class is algorithms) s
5 665 M
(     that, while promising, are not deemed safe enough for deployment on) s
5 654 M
(     the Internet, but are being specified to facilitate investigations) s
5 643 M
(     via simulation and testbeds.) s
5 621 M
(     Each alternate congestion control algorithm published is required to) s
5 610 M
(     include a statement in the abstract indicating whether or not the) s
5 599 M
(     proposal is considered safe for use on the Internet.  Each alternate) s
5 588 M
(     congestion control algorithm published is also required to include a) s
5 577 M
(     statement in the abstract describing environments where the protocol) s
5 566 M
(     is not recommended for deployment even though it may not be unsafe) s
5 555 M
(     for use.) s
5 533 M
(     As examples of such statements, [RFC3649] specifying HighSpeed TCP) s
5 522 M
(     includes a statement in the abstract stating that the proposal is) s
5 511 M
(     Experimental, but may be deployed in the current Internet.  In) s
5 500 M
(     contrast, the Quick-Start document [QuickStart] includes a paragraph) s
5 489 M
(     in the abstract stating the the mechanism is only being proposed for) s
5 478 M
(     controlled environments.  The abstract specifies environments where) s
5 467 M
(     the Quick-Start request would give false positives \(and therefore) s
5 456 M
(     would be unsafe to deploy\).  The abstract also specifies) s
5 445 M
(     environments where packets containing the Quick-Start request could) s
5 434 M
(     be dropped in the network; in such an environment, Quick-Start would) s
5 423 M
(     not be unsafe to deploy, but deployment would still not be) s
5 412 M
(     recommended because it could cause unnecessary delays for the) s
5 401 M
(     connections attempting to use Quick-Start.) s
5 379 M
(     For researchers who are not ready to bring their congestion control) s
5 368 M
(     mechanisms to the IETF for standardization \(either as Experimental) s
5 357 M
(     or as Proposed Standard\), one possibility would be to submit an) s
5 346 M
(     internet-draft that documents the alternate congestion control) s
5 335 M
(     mechanism for the benefit of the IETF and IRTF communities.  This) s
5 324 M
(     is particularly encouraged in order to get algorithm specifications) s
5 313 M
(     widely disseminated to facilitate further research.  Such an) s
5 302 M
(     internet-draft could be submitted to be considered as Informational,) s
5 291 M
(     as a first step in the process towards standardization.  Such a) s
5 280 M
(     document would also be expected to carry an explicit warning against) s
5 269 M
(     using the scheme in the global Internet.) s
5 247 M
(3.  Guidelines) s
5 225 M
(     As noted above, authors are expected to do a well-rounded) s
5 214 M
(     evaluations of the pros and cons of proposals brought to the IETF.) s
5 203 M
(     The following are guidelines to help authors and the IETF community.) s
5 192 M
(     Concerns that fall outside the scope of these guidelines are) s
5 181 M
(     certainly possible and these guidelines should not be used as a) s
5 170 M
(     check-list.) s
5 148 M
(     \(1\) Fairness to Standard TCP, SCTP, and DCCP.) s
5 126 M
(         In environments where standard congestion control is able to) s
5 115 M
(         make reasonable use of the available bandwidth the proposed) s
5 93 M
(Expires: June 2006                                              [Page 3]) s
_R
S
%%Page: (4) 4
%%BeginPageSetup
_S
24 24 translate
/pagenum 4 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
5 720 M
(draft-floyd-tsvwg-cc-alt-00.txt                            December 2006) s
5 698 M
(         change should not significantly change this state.) s
5 676 M
(         For instance, in a situation where each of N flows uses 1/N of) s
5 665 M
(         the network capacity, a new congestion control scheme should not) s
5 654 M
(         significantly deviate from this state.  For instance, a flow) s
5 643 M
(         using an alternate congestion controller that took half the) s
5 632 M
(         capacity and left each of the remaining N flows with 1/2N of the) s
5 621 M
(         capacity would be suspect.) s
5 599 M
(         We note that this bullet is not a requirement for strict) s
5 588 M
(         TCP-friendliness as a prerequisite for an alternate congestion) s
5 577 M
(         control mechanism to advance to Experimental.  As an example,) s
5 566 M
(         HighSpeed TCP is a congestion control mechanism that is) s
5 555 M
(         Experimental, but that is not TCP-friendly in all environments.) s
5 533 M
(     \(2\) Using Spare Capacity.) s
5 511 M
(         Similar to point \(1\), alternate congestion control algorithms) s
5 500 M
(         may take up spare capacity in the network, but may not steal) s
5 489 M
(         significant amounts of capacity from flows using currently) s
5 478 M
(         standardized congestion control.) s
5 456 M
(         For instance, consider the case where N flows cannot naturally) s
5 445 M
(         use all the capacity and equally share one-third of the capacity) s
5 434 M
(         \(with each flow using 1/3N of the capacity\).  A single flow) s
5 423 M
(         using an alternate congestion control scheme could use the) s
5 412 M
(         unused two-thirds of capacity.  However, the flow using) s
5 401 M
(         alternate congestion control should not steal significant) s
5 390 M
(         amounts of additional capacity from the N standard flows.) s
5 368 M
(     \(3\) Difficult Environments.) s
5 346 M
(         An assessment of proposed algorithms in difficult environments) s
5 335 M
(         such as paths containing wireless links and paths with) s
5 324 M
(         reverse-path congestion.  In addition, proposed algorithms) s
5 313 M
(         should be evaluated in situations where the bottleneck has high) s
5 302 M
(         and low levels of statistical multiplexing.) s
5 280 M
(     \(4\) Investigating a Range of Environments.) s
5 258 M
(         Similar to the last criteria, proposed alternate congestion) s
5 247 M
(         controllers should be assessed in a range of environments.  For) s
5 236 M
(         instance, proposals should be investigated across a range of) s
5 225 M
(         bandwidths and round-trip times.  A particularly important) s
5 214 M
(         aspect of evaluating a proposal for standardization is in) s
5 203 M
(         understanding where the algorithm breaks down.  Therefore,) s
5 192 M
(         particular attention should be paid to extending the) s
5 181 M
(         investigation into areas where the proposal does not perform) s
5 170 M
(         well.) s
5 148 M
(     \(5\) Protection Against Congestion Collapse.) s
5 126 M
(         The alternate congestion control mechanism should either) s
5 115 M
(         stop sending when the packet drop rate exceeds some threshold) s
5 93 M
(Expires: June 2006                                              [Page 4]) s
_R
S
%%Page: (5) 5
%%BeginPageSetup
_S
24 24 translate
/pagenum 5 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
5 720 M
(draft-floyd-tsvwg-cc-alt-00.txt                            December 2006) s
5 698 M
(         [RFC3714], or should include some notion of "full backoff".) s
5 687 M
(         For "full backoff", at some point the algorithm would) s
5 676 M
(         reduce the sending rate to one packet per round-trip time) s
5 665 M
(         and then exponentially backoff the time between single) s
5 654 M
(         packet transmissions if congestion persists.  Exactly when) s
5 643 M
(         either "full backoff" or a pause in sending comes into play) s
5 632 M
(         will be algorithm-specific.  However, this requirement is) s
5 621 M
(         crucial to protect the network in times of extreme congestion.) s
5 599 M
(         If "full backoff" is used, this bullet does not require) s
5 588 M
(         that the full backoff mechanism must be identical to that) s
5 577 M
(         of TCP.  As an example, this bullet does not preclude) s
5 566 M
(         full backoff mechanisms that would give flows with) s
5 555 M
(         different round-trip times comparable bandwidth during) s
5 544 M
(         backoff.) s
5 522 M
(     \(6\) Fairness within the Alternate Congestion Control Algorithm.) s
5 500 M
(         In environments with multiple competing flows using the) s
5 489 M
(         alternate congestion control algorithm, the proposal should) s
5 478 M
(         explore how bandwidth is shared among the competing flows.) s
5 456 M
(     \(7\) Performance with Misbehaving Nodes and Outside Attackers.) s
5 434 M
(         The proposal should explore how the alternate congestion control) s
5 423 M
(         mechanism performs with misbehaving senders, receivers, or) s
5 412 M
(         routers.  In addition, the proposal should explore how the) s
5 401 M
(         alternate congestion control mechanism performs with outside) s
5 390 M
(         attackers.  This can be particularly important for congestion) s
5 379 M
(         control mechanisms that involve explicit feedback from routers) s
5 368 M
(         along the path.) s
5 346 M
(     \(8\) Responses to Sudden or Transient Events) s
5 324 M
(         The proposal should consider how the alternate congestion) s
5 313 M
(         control mechanism would perform in the presence of transient) s
5 302 M
(         events such as sudden congestion, a routing change, or a) s
5 291 M
(         mobility event.) s
5 269 M
(     \(9\) Incremental Deployment.) s
5 247 M
(         The proposal should discuss whether the alternate congestion) s
5 236 M
(         control mechanism allows for incremental deployment in the) s
5 225 M
(         targeted environment.  If the alternate congestion control) s
5 214 M
(         mechanism is intended only for specific environments, the) s
5 203 M
(         proposal should consider how this intention is to be) s
5 192 M
(         carried out.) s
5 170 M
(4.  Conclusions) s
5 148 M
(     This document is intended as a guideline for researchers) s
5 137 M
(     in bringing congestion control mechanisms to the IETF to) s
5 126 M
(     be considered for Experimental status, and also as a) s
5 115 M
(     guideline to the IETF in evaluating such proposals.) s
5 93 M
(Expires: June 2006                                              [Page 5]) s
_R
S
%%Page: (6) 6
%%BeginPageSetup
_S
24 24 translate
/pagenum 6 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
5 720 M
(draft-floyd-tsvwg-cc-alt-00.txt                            December 2006) s
5 687 M
(5.  Security Considerations) s
5 665 M
(     This document does not represent a change to any aspect of the) s
5 654 M
(     TCP/IP protocol suite and therefore does not directly impact) s
5 643 M
(     Internet security.  The implementation of various facets of the) s
5 632 M
(     Internet's current congestion control algorithms do have security) s
5 621 M
(     implications \(e.g., as outlined in [RFC2581]\).  Alternate congestion) s
5 610 M
(     control schemes should be mindful of such pitfalls, as well.) s
5 588 M
(6.  IANA Considerations) s
5 566 M
(     This document does not require any IANA action.) s
5 544 M
(Acknowledgments) s
5 522 M
(     Discussions with Lars Eggert and Aaron Falk seeded this document.) s
5 511 M
(     Thanks to Colin Perkins for feedback.  This document also draws) s
5 500 M
(     from [Metrics].  Thanks to Colin Perkins for feedback.) s
5 478 M
(Normative References) s
5 456 M
(Informative References) s
5 434 M
(     [RFC2581] M. Allman, V. Paxson, and W. Stevens, TCP Congestion) s
5 423 M
(     Control, RFC 2581, Proposed Standard, April 1999.) s
5 401 M
(     [RFC2914] S. Floyd, Congestion Control Principles, RFC 2914, Best) s
5 390 M
(     Current Practice, September 2000.) s
5 368 M
(     [RFC3649] S. Floyd, HighSpeed TCP for Large Congestion Windows,) s
5 357 M
(     RFC 3649, September 2003.) s
5 335 M
(     [RFC3714] S. Floyd and J. Kempf, IAB Concerns Regarding Congestion) s
5 324 M
(     Control for Voice Traffic in the Internet, RFC 3714, March 2004.) s
5 302 M
(     [Metrics] S. Floyd, Metrics for the Evaluation of Congestion) s
5 291 M
(     Control Mechanisms.  Internet-draft draft-irtf-tmrg-metrics-04,) s
5 280 M
(     work in progress, August 2006.) s
5 258 M
(     [QuickStart] S. Floyd, M. Allman, A. Jain, and P. Sarolahti,) s
5 247 M
(     Quick-Start for TCP and IP.  Internet-draft) s
5 236 M
(     draft-ietf-tsvwg-quickstart-07.txt, work in progress, October) s
5 225 M
(     2006.  Approved for Experimental.) s
5 203 M
(Authors' Addresses) s
5 181 M
(     Sally Floyd) s
5 170 M
(     Phone: +1 \(510\) 666-2989) s
5 159 M
(     ICIR \(ICSI Center for Internet Research\)) s
5 148 M
(     Email: floyd@icir.org) s
5 137 M
(     URL: http://www.icir.org/floyd/) s
5 115 M
(     Mark Allman) s
5 93 M
(Expires: June 2006                                              [Page 6]) s
_R
S
%%Page: (7) 7
%%BeginPageSetup
_S
24 24 translate
/pagenum 7 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
5 720 M
(draft-floyd-tsvwg-cc-alt-00.txt                            December 2006) s
5 698 M
(     ICSI Center for Internet Research) s
5 687 M
(     1947 Center Street, Suite 600) s
5 676 M
(     Berkeley, CA 94704-1198) s
5 665 M
(     Phone: \(440\) 235-1792) s
5 654 M
(     Email: mallman@icir.org) s
5 643 M
(     URL: http://www.icir.org/mallman/) s
5 621 M
(Intellectual Property Statement) s
5 599 M
(     The IETF takes no position regarding the validity or scope of any) s
5 588 M
(     Intellectual Property Rights or other rights that might be claimed) s
5 577 M
(     to pertain to the implementation or use of the technology described) s
5 566 M
(     in this document or the extent to which any license under such) s
5 555 M
(     rights might or might not be available; nor does it represent that) s
5 544 M
(     it has made any independent effort to identify any such rights.) s
5 533 M
(     Information on the procedures with respect to rights in RFC) s
5 522 M
(     documents can be found in BCP 78 and BCP 79.) s
5 500 M
(     Copies of IPR disclosures made to the IETF Secretariat and any) s
5 489 M
(     assurances of licenses to be made available, or the result of an) s
5 478 M
(     attempt made to obtain a general license or permission for the use) s
5 467 M
(     of such proprietary rights by implementers or users of this) s
5 456 M
(     specification can be obtained from the IETF on-line IPR repository) s
5 445 M
(     at http://www.ietf.org/ipr.) s
5 423 M
(     The IETF invites any interested party to bring to its attention any) s
5 412 M
(     copyrights, patents or patent applications, or other proprietary) s
5 401 M
(     rights that may cover technology that may be required to implement) s
5 390 M
(     this standard.  Please address the information to the IETF at) s
5 379 M
(     ietf-ipr@ietf.org.) s
5 357 M
(Disclaimer of Validity) s
5 335 M
(     This document and the information contained herein are provided on) s
5 324 M
(     an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE) s
5 313 M
(     REPRESENTS OR IS SPONSORED BY \(IF ANY\), THE INTERNET SOCIETY AND THE) s
5 302 M
(     INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR) s
5 291 M
(     IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF) s
5 280 M
(     THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED) s
5 269 M
(     WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.) s
5 247 M
(Copyright Statement) s
5 225 M
(     Copyright \(C\) The Internet Society \(2006\).  This document is subject) s
5 214 M
(     to the rights, licenses and restrictions contained in BCP 78, and) s
5 203 M
(     except as set forth therein, the authors retain all their rights.) s
5 181 M
(Acknowledgment) s
5 159 M
(     Funding for the RFC Editor function is currently provided by the) s
5 148 M
(     Internet Society.) s
5 93 M
(Expires: June 2006                                              [Page 7]) s
_R
S
%%Page: (8) 8
%%BeginPageSetup
_S
24 24 translate
/pagenum 8 def
/fname (draft-floyd-tsvwg-cc-alt-00.txt) def
/fdir () def
/ftail (draft-floyd-tsvwg-cc-alt-00.txt) def
/user_header_p false def
%%EndPageSetup
_R
S
%%Trailer
%%Pages: 8
%%DocumentNeededResources: font Courier-Bold Courier 
%%EOF

PAFTECH AB 2003-20262026-04-22 17:01:27