ࡱ> y 1  !"#$%&'()*+,-./023456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxz{|}~RdO)nβPictures KPowerPoint Document(SummaryInformation(tL ( / 00DArial ProN W3ڄnlD0D-3 00000oN W3ڄnlD0 DWingdingsN W3ڄnlD00D0000҉0 ProN W3ڄnlD0 ` .  @n?" dd@  @@`` z43        !"#$%&'()*+,-./0123?b$ ە/+h;K_b$XO w/\c b$`B;;˃ i;S @8y]uʚ;2Nʚ;g4UdUdCмbppp@ <4!d!d0ׄgʚ;<4dddd0ׄgʚ; :\___PPT10  H P (P HH`   0 8(P 0 ( 0H 8(0N ___PPT90 ( @@  Pn ە/+h;K_PNG  IHDR, 9!sRGBgAMA7 cHRMz&u0`:pQ<fPLTEf3333ff3fff33f3333333f3f3fff33̙33̙3f̙f3ff3f33333F p+tRNS'4 5 cmPPJCmp0712|4IDATH E#ZSkTdTc/@:CDi/G&HHw˫ ZmUս>@ b1ȑc1^~{~~w1~DI8DXoc}$)E$UU;{G >3cE)̙ͩ,g0@ݬy:㸐/yz'''P8W_%5YUUkt9zHvH:oJtt Xgn]zZ=^PU>>??O3sOFNIsZ\x21n0]LvM On{NSU[mUG+;okVV$H@o"" ì/v+CҬ$5F q0"Sd;G&I}e⒈a~ɉo/mmH wF;KjwEYcqۡLRU%ޭi".eXN"+UH_M2efyMjSR2[>gd=s],3;a:du먉wkrbc*Ih~|Bȑ9e]9d\*RBR9s׹$sued7fyj+g 5]U'QaEުfVȐnm 2xf?_f]E({3p9`sN|Eie\IP`T5 $G&djdsN͔}Ձ@3n=WXIݚ&Z39;'U@YKRzWL1O" hM:L]7ku5ެK]Rn~ 9UV AJ4 (/MSJ)=+:go b=2#Z&ȐgLM x9O*rN7U$nj!63I$M?wO>'o!$DIENDB`nz XO w/\cPNG  IHDRX _a;gAMAPLTEff̙f̙3̙f3f33f3f3333ff3fff33f3f3f3f33f33333333333*pMsڇa QQEܩqN|]+5 ~psaauXru4%UŎۚ9tRNS AbKGDIDATxXr6 tƦrG(RdVu% /7鄒%QI`XwnΡ77RHc")46&smIO~k[qۍݚpodct>u`.i٥ÞZeV1&[a^+ruՖcǽ{e{u.s: *|4kVfo7u q" 7 m 3P rC~nӎ7|DB&(#D ʕtv^\C*ƭ&)\*OAɵo% ڧW X?4R=liB`|U؉f3PLi43JhA*|>;T ^G*w-njY590'gX|~MEW!-]kdIsY8[g8|76f%zM;4֢k9?%uJA#Rf+K**[ݰ&j$̬w4bF^wm\Fv81'qO!ɩ Wbu, X2هb97KImTH,IvDUgqIojZ]^}輥%>rJY2gBvk2j[o6 w̼Ƭk|6dss{-:frfhazfƔz݆j:Q摀NaR#đW$.1>J(a״,k&dm_c+mJ܏7!Eu[+ D?#_Hq$ U(*osDffrora|s x$u%9d\cz ^z\* ~5&%h820"Ԅd^>U*um\C@r&FVTdof-}f>QW%:嗔+2TZ䓅er JKTq^mC3Hz¢Q;ӠA4Gy,>`)<9, *um(l EmJd#QWݯ` X"dZhx.T,hWOt &fT)[B͹o(e- j^sE+H+#鬐0,w,P_3xAjXK/u u$RZU"?1R>YW4e҆$%ɮLIJ뱵8bUpT_G ɞeAbѽ2' E[uʦ#UРWO:ݕV1"^t2MV<6{QdRmZ(Sѻrˌ0Nk #HQHIGS r\,.5sů'&XkFJ[&3/z0D,G$CBP/=MLjOBni6*Q {FY+ =\Di"aLf-ER2t /@w?7oiVUsFIENDB`na;`B;;˃ PNG  IHDRX<B4sRGBgAMA7 cHRMz&u0`:pQ<PLTE33f3ff3PP̙f3fff333f33f333f33f333f333f333ff333ff3ff3f̙3f̙f3f3f3ff33f3f33f33ff33ff33fff̙f333tRNS@f cmPPJCmp0712g9IDATx}kwȒڽ=uے%B{U(16#" 3CR$E Ȭ]K1Dܦ)NM]wlƧ]&64 ~t0a\o~i?Gӭyhm'MLa]um .V6]ǩ>/`6ڵio-=p[>tB .qL"~ojfׂzi)}Χ=~>‡_q?ys{yRJ]tN//'罽l?=ϓ/×߿?`O;hkz =s>rܦ)|a˗x`_s?~^mv][a!9]n17DZmqoci O_/__/_=|1}^ @a+L@]gu!g{E|9}9 tԙ' xyq2xlprWc{d,P,; ߇_ ?qўw_|&{T17\rO_&·bbq&27c]04s35g,˶#hyZa|.PnM44MX5(A 3+ii0*a}ZC s6xN0Vͻ]+cPpJ^Xӡ3/ Sy/ Wϟdzׯvk,nw$#7l?xN8mٴe,aˠΚ~ǂ 5U2k@>>|؜@{ kf2"=Yxɇs2dxV3:j[Lq<ě1,CX{eAˮdoloA8+CUζd8YT4:/{O`==#}<aBovŭǧO_^>+C!l_N(b,G>۝=ʾXycXae='g,Ba(en%raq͢!$? I  8O!0vHdkD#iʘST#Ƙ 3$a>2 +F-c v+o3mtMC.@Su@Q`%{cQ\>~i,i=cy6Xr!=BY2,fx|_p|748_J:d2+&1oX/2Hx/ `BBξ8p-Jg;%dX8Y$,@k'G^Rke-q>6? (>>4qV0c DTT_V&Z@A͎AY@Vs2.V캨[4eȌ@eYT`-&,C!VfUk>+er`V!G$;SZfCƲ#`xt (;!~~/7G'bW~d[/P:T+Iv?/GCgX<.gJVa D$9U !aB ~Z0Xh3#k2\ wFPRp0k+0LbN8ktdgrؕ)%Vhm02u% 7FpF428#υ |thgOqX"f=ak~,@k\*bfkg7y[`Mz[2Ͷ@C ؚrcq+SxT/^^pL& qVq<;eGC0Y;p6bY1hw#ʬQeS LBHw6"~! 큪'&v !xbpE$<ЎY9zKAYWG(: )@2AǙD%EK]23@0A,1(w;'#1YaWw")m] L)" PZĄMaeQȈvю: +0mYP(3:QV ,hۅ`J ZЊ]@ޗYLt/R^FmeigUcjeȗLn/Ҟg8/~Y@X%zFY~dNO?l_xRG4ҁ ZH؉KZ*G"EhK(h+\ La])9L,4Pbv!ɧ@qBMVs$lvgy'!˖p*,> < R.>ҮkO!3 -۳ʯB %.!515,1> #4Yaa2bL;S$-ՂV$!ap,.Bd(TA ˂-IP]j¢V  q8ar^wد)8d>ەIߦ%Z7!Zz- ]KvfGblq)tD4Oʞu{3dbV 0|va 4P,z22d=1踕פ Qǎ"dc(!h1@7Nf%daGwrNw.@Uuo S$<)l'Beo 3z 8&w }j !H5W!Xv&[hfzH7dH֌"c@c:vZG (!\Ԝs萑)-Z8mYl5m IV8Z|e8-&@eY{I`۷$(54 U *%,>^>=ݑO_ T}ʐ2~}Qۆ#)yP, N㱡T; pwVV4jA P3Ʋ\cP[GTݷ QPߴˉHP5+l`T - 9SR˿*cim%c wmtXELD$Z2L߽$jqp>_"пAtxi #4Eir\ž 1kymiW~jh +I!Ϣ2FhqVEO'G :z 円@FB }O8BVO?(®4Yx"ͦNd>% !Uxt*;FU23 e1XaB=!Ryp,!i{& ~`trHEʇ@.DX'7Zg)'(9eXby Ŗ&`Vj눬X/\H;ʡ1jl!-bhB'˦LR^aasY0(Nΰ1\o{EDl`P mt׺†P_>$Oi8YGXl"b 07b1H5"g2!2R2gֽ,DRT3 ?S؋;8!B*,4dzIb@5:Z΄ɠ^4y  )ɫ'3\Y&JaogMpĠTF&ITKWDU*ZNpn2~C~w "턝0PPu v};e̱rY2Y5B@! s >Zk Rk0ckKF<'؎ DV"kG=K0⫕jj/egHYbD05*KSk'!P2 2<N#l5v@jᖠOe{G8J f9}8xV'k++F!#3~UxBQŴ_ee%$dcp$Si.副 (2zRF[GZϴ0¹#pyap&:\ƞz!C8L?'@u D35>: k3 1ޢ7+ŁqhP0G(BU\dW"GAnșt$J'Wr bŎtI f:  @];a,a}\q< ܜ&xx}b!C6IIA˺ .a3 P/=dl'Ϸ`F@ U nK˽`Sd;`+ 4C/ޠϡ֥YbՃ4RbG)z#ڍ #_ '0VYDR.[HRaQ/"K~OH DU {\IQoߑC\x#Ri?X#8<[;)^]rD"ҊmвC֏*PYn*JFVCh'K]d,Tu#aH7-\*& 朴վ`)?kN=6Gب$]:2KKB(X+/Q=medJ8]dRl3J?KX=feTUP(MZIU&Be1H%*ϕoOހt!8g F>+=2)dFQ\Զy*O|ȇFHd},-r)aC`z"%DT Ooi6([r0hW Kg5:c:{YJ䀫'gQQ2SljkG 6m޶~`XTR_G DKL0Vlo|4 MK 2\=W+ sr\z:]NW}.?w`d3%m o7j/&2Zc-K!#J|ъ, T SO /ѹ"M*HYBY1NTTA%]nl=ղdx \0$5)i.3x?_dAYSSAvƫ) h%Z(ʂ*,h`ߥaGjONf)I]%Sn{COڈKJNYW#71p/ 2Nu^(FkzY (x+$eBHX۴i͗'7APY2߁jaJ^T0+'ɤx1~nG*;:NƈB@≃;@.xgb*dՂR:I13nUY`󑕖(T\Tc + *AnܶI&v7wqlEej$ȏ*TөEV4+ jlEY4H]fm`])2?蘆E[56"~`cetGW JB# >ufmԱڨQv )w""r_q{I-|hM #[AU"m]~bd GyXL އ"@LV ".bJ?ﮕ6L&{4%*( }ȫhM%fA#'Od'jGB LLR3yQÈ!kU3#CI&3U,q0Y؟ q1Φz' JaLRulrX<']!]Wynư}Y*d$U(WBJCRJF[`ڼ)UkZyaY4JAUhIcC: Z!uנYv溅Igo@:D)A2~ݹzwxck=#+}//c݉@f֐aQCV0O: UY&Fc5*3 ?g_|F`v}RJQr c  ]$qBYD\7˄i]BdʑUb,*nF[+ Ejoopy L;9]6haZk~ӝ[^)#'GҘ@RQS[Qq؎Rd$eQd:U!U^M E"lzy9UUVr=Uni1D*(^= Ғ"\e-F^ey>8,ǚxTmMlw oU ]Yߗׅc-4Z䦎TN y|䐲OhцO? O;;LEɔ ޒ W V192 kEed#-=(Yd<ScܥWW6_V_].m#-$:m'?HPatLX5$nNozx?cuEҒ"ۗ2aCMwnrHoHMPkmum#baZ\5ʱ0"6 A wuq@sl6o0Xg.N?h" KlЇwD ӛ'%Z ϩTNW-%  q`FXgUO-I<(Ȱ Mg/ &.IgrW"okqX,)I7u06.WjuX0߮V{(YZN|o\a޽\X+aCU!Em:UϾ饭`8ՏX#"pZʩS{QӯЈ#WZ/b5ҎnOwH#a,C^%JW D4J/@=ِçԊ!F UW ,xkTYTZr< Z^ȅ #L/KQQgϐ>- =vJ)ÂG0 .ۢ IZ^5j%8"ow ] krdŵ'b ;GVG"txl K3&Dӵrjv0-2G!u-$IbݛRڕ~V0JuUWAPO;𢶭"X gP_ tMTz)r$V$NÒ?P;\/X1 B.W'v,VQ_ӿ2Gr YHޛ欼E-a& ɦpۊv{k• 1MmܖĪ*t,չDe>"DHthy;N-n68:dF|FŠV!E%QGQP>v7XW v xX^Rِ%,Z31Å p˾L?NXoX aa(O+>J\3:R1U5T{m3U;M!&uzM\|Po65 kY̌Ik֫n}iRE._5pEZ ; E4ƒVwPAjZV\[lQj1Z~4I3]'%1xhCp2&U8f&Dt*3,o8k@%W5mps%k߷i rM Xq 7~htGϠCq f y<N9MI5K~^+2;xiF.@Ճ6B\_`_ H 91ĦAvڇ8UiIŵ{ƍ5+#vƯ'MMY"eF"Pk ܑ?v2 w\K1A7@fM߿IM H<{)˙2dgWùʶYһNWaiʠ_7_ءuQ)_D[mP9:,m<}c}4 ?u0r^zgDP[IF_=Պ!6M CIG smCJV>-V¡e}CQ:t#M^|>kV )8λ#zW~7[;|) :諶TA\,a Y}SLJcRa;r\$ZTk6EaW!'NYo4ɾx 5hC0x럧7ZJFoqiiN`55;ƒƊgPy]Denx`jm&]9OӭQ07YcMWqRT[Gw rT|TTP\&Dwh_`{IIJN XFVaiRWE?RGWZ1X+,j:L23fQ4@HW-ezzܫ\3D:>BYl)O)o @Z?FH8z,D? $vPcJlv;s/Vy2@ʽu?i)鍲OU~8W vLaQTu[s1bTTNyL"cε6jSCv |{*/ C|:.1w4C^mV?\+΂*N[. 28oA [ncIiXm\z{2VsF ! nC4i"5Sir{'e"=7i;uUmAdWG1\+ӹ"zA04dZo.bZ)j,2H$CE]z\h3n4]Tc8)2?<\!N߼EK{MNֻHIjY:)< LU~M6}Z+VO!n <*k\wVN`GHcx ow}Bx#:ol)‚Nn3,Q\N3 tD+:̱2V#flpNBܜTӾ ʶpy0o+66v, dHACt%WGٯZ #f)^ʩc(zi(xͅu~Z{CU|vĝ] G~ ?Nuwpn:h,i-8-)a鈩P-i#$Kil7iI;5co'[Jo߹ߕ;S]_Zc4Lgo[&$>b: C/%yꞭg_AoY "UsN^VNYWH{pظ )~:iW7(. V ͬooɛ& ≔V pʏbmQUG- @gpUmgEx{gVN+nN|0 <3Xmq~:eNeMTY$dDŽfUi0n^ 5'eLтb*j'і5>-3"6K eUf+"∑NBjJԅ.ƺfvqh@cz[†?AaiI Kތwݧv g__VONV .kCmzwHI#nPwQy5Vm=^)7ژ[-]|zz?c.Oxv"fs=&W.@c0H#p3&+Roۙ!i5'K? i$d-RpC _fY▕u{4ws:b&8lGo/x6magSˇ#k3Gj} :]&/RJ>L>!@VXG.*v`oozpLYj8߅u'$9>sal3z۽d,>טhk,IENDB`    ` pP 0` P ` pP`h___PPT2001D<4XI___PPTMac11I@f   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography B0   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  0   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  d   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  h   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography $   hnamd` Arial&Monotype Typography  h   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography D   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography 8   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography   ,   hnamd` Arial&Monotype Typography $   hnamd` Arial&Monotype Typography  h   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography ,   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  8   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  $   hnamd` Arial&Monotype Typography   8   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography   ,   hnamd` Arial&Monotype Typography  4   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography H   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography ? %HFIND experimental requirementsDavid D. ClarkFIND%Future Internet Design (FIND) is an NSF program (now folded in to NetSE) to envision the Internet of 15 years from now. FIND was one of the early drivers for the development of GENI. FIND research spans the layers from technology through classic Internet architecture to application design. 0&y@l >B0>mThe original Science PlanBAs part of the original work on GENI, we prepared a Science Plan (also called the Research Plan), which listed various requirements. Those requirements have not changed. The GENI plan, as then proposed, (and now?) did not fully meet those requirements. Suggests a need to revisit some assumptions and design approaches.>ZCZ&TC Pg$RCGeneral requirementsOA real, distributed network. Not a bunch of routers in a rack. As wide a reach as possible. Reach to the edge. Allow experimental edge equipment direct connection to the GENI-based experiments. Access to real users. Creates a tension with the desire for realistic lower layer technology, e.g. optical layer. Long-running experiments. ?S\?S  \wBackground commentaryThere are a wide range of experiments (with a wide range of requirements) that might be posed for GENI. It is not realistic to imagine that we can build a single fully general facility that can support all of these experiments at reasonable cost. Implies a need to be creative, make clever choices. Make some compromises. The MREFC process distorted the translation of requirements into facilities E.g. a single unified facility. All of these assumptions should be reconsidered. One or many? Is GENI one thing? A better set of outcomes?ZKZLZ Z1Z:ZhKL   1   PIQB - The excitement is at the edgeWhile (some of us) want to mess with the core Security, management, etc. The real action is at the edge. New devices (mobile, embedded) New networks (wireless of all sorts) Not all parts of the network will look like ethernet! Cars and other mobile networks. To emulate the future, need all this in the experiment. Not so clear how to virtualize. Does this matter? Remember the goal of reaching real users. /ZZ ZDZ6Z!Z8Z]Z/   %6!82  +$$- S  ="The excitement is at higher layersDesign patterns for applications. Highly distributed, clouds, etc. New support mechanisms Identity frameworks, location frameworks, etc. Important to ask: to what extent can we explore these on the existing Internet?b""/P""/ P New protocol stacksResearchers want to try new protocols at the network, transport (and higher) layers. New means of authentication. New mechanisms to deal with soft state in the network. & Implies the need to replace the protocol stack in the end node. Facilitating this should be part of GENI effort. Mobile devices, not just PCs. Remember, we want real users. A real experimental tension here& VZVZAZZV7  A1  "$$,TkLower level research?Two sorts of reasons. We need to do a better job supporting apps. Security, availability, management, economics, etc. (It is not the data plane, except indirectly& ) What happens there will influence the design of applications. The research is not independent. We don t have the lower levels right. Management, operations, etc. Some of this is perhaps more independent.Z,ZZ!Z&ZHZ,4. >!&H Pb* PacketsWe want to try out packet formats that are not IPv4 and IPv6. Congestion and its control. Novel addressing modes New mechanisms for security. New concepts in network management. New schemes for tracking payment. & This capability must reach all parts of the experiment. >:>  $"9   DNew functions  in or  on the netNot all boxes that are topologically  in the network are routers. Security enforcement devices. Encryption devices. Packet inspection devices. Application support devices. & Implies different node requirements. More storage, processing, etc. Very high performance. Network devices today are highly purpose-tuned. How can we provide generality for a range of experiments with different topological requirements? Processing nodes should be in the net, not just at the edges.CZlZ%ZZC  %g  b$$=((,,b$~;F Emulating a real networkThe core of a big ISP today does not forward packets, but is built of flows that carry aggregates of packets. Optical lambdas, MPLS circuits, etc. Not all parts of the network will look like ethernet! Cost and complexity is a major driver in real nets. A future architecture should do a better job of linking the management of these layers. How should these capabilities be made available to experimenters? Should this be a GENI goal?jnZ\ZZn\5X  ]b$3$, Optics in the coreThe original proposal for the GENI platform had rather sophisticated optical components in the core. E.g. ROADMs. This had major cost implications. This had major  virtualization implications. How do you virtualize a ROADM, since it is not a packet device? Not well done in original proposal. The goal was to at least emulate how real networks look today. eZ ZPZdZ@Ze P@  #@P/:2* One alternative`Let the folks who want to play with real optics build their own environment. Smaller scale? Build the large scale testbed out of packets (IP, ethernet: does it matter?) and tunnel our  new packets inside them. Is this a better idea? (It has limitations that should be recognized, as well as benefits.) bNZZZZNw[ bU=>Picking compromisesIf we fold optics in& More realistic (but for what class of experiments?) QoS, non-packet end-to-end services. Intrinsic availability, security, management, etc. Cross-layer protocol designs. If we use packets and tunnel& Much easier to achieve scale and reach real users. Lower layer  technology complexity will have to be simulated. Is this an issue?4v4%3  3P  $$JAnother way of saying thisThe independence of the parts of the Internet (e.g. apps from link technology) is as a result of the  hourglass design, the end-to-end design, etc. Assuming that experiments can tunnel over packets risks baking in today s hour-glass architecture in tomorrow s experiments. My opinion: the risk of guiding research toward a presumption of the hourglass has to be mitigated in some way.6p}pp  ManagementOAs the previous slide tried to emphasize, a lot of future Internet research is about management, not the data plane. So GENI is not just virtualized data planes, but virtualized network management schemes. Fault diagnosis. Virtualization messes this up. Set up and tear down of circuits. Want to mimic real operators, not just users.jZZZPZvZ ".bE0  A general challenge for GENIMany of the proposed ideas for a future Internet stress management, security, etc. It may be less clear how to virtualize the experimental infrastructure to allow these to be demonstrated than to virtualize the data plane. For example, to demonstrate improved availability, the GENI platform would ideally mimic the baseline failure modes of the eventual technology. What is the best set of compromises? Again: one GENI facility or many?b%"S%  "t0! "'  Scale and speedSome experiments stress scale 100 s of routers, 100 s of fixed end-nodes, N00 s of mobile devices, and rich connectivity. Others mention  thousands of interconnected devices . Speed was not an issue for this experiment. These are miniscule experiments! d!\5 ," Real scale10K-100K AS regions. Routers with 500 10 Gb ports. 100 striped lambdas& Millions of multi-homed end-nodes. Highly heterogeneous environment. How are we going to try (and validate) ideas at these scales? 3ZZEZZ?Z# "= b! 2[ Some specific requirementscRich connectivity (to experiment with novel routing.) Need for multiple regions (emulate different operators). High bandwidth paths between slices? Heavy-duty isolation among slices. Detecting physical location. All devices should be able to do this. Universal crypto capability. Allow experiments in virtualized architectures. Recursive virtualization?oZ%ZAZ(ZMZZ69%A  (MPn JInstrumentationNIn some respects, a confused discussion. Clearly, we need to gather data on how experiments are going. But is this done in the infrastructure, or in the slice? A future Internet must include native capabilities for instrumentation. Is  data something generic that is shared? Wishful thinking?Ng>)>:I >P(A4G>Non-requirementsSatellite Not a sufficiently distinct challenge. Residential access networks Again, a challenge, but not sufficiently distinctive. If have wireless nets and high-bandwidth links to end-nodes, good enough. x '6J '6 JuXThe experimental landscapeHow many experiment? PlanetLab has demonstrated that there can be 1000 s of experiments. But how many folks want to try out a new Internet? Perhaps we need different sorts of tools for experimental deployment at different  layers . Again, one GENI or many?XD4\ , lu/ !"#$%&'() * + , - ./01234567 `0` MMM̙f` f3 ޚ` ` ff,,B3f33` P3fPP3` ff333tfff` @EoOV>?" dd@,?nF3d@ `A  @`K``A n?" dd@   @@``PR    @ ` `p>>    X (   BT ~d   "`n   S 0AARTHSEPA"5~dh   C *AArthsepa" D$  0G "`  RClick to edit Master text styles Second level Third level Fourth level Fifth level!     S  6Ps "s#  0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  F* b  6pD "(  0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  F* b  6u "\  0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  B* `  S C "`0  T Click to edit Master title style! !T  <d޽h ? MMM̙f Future Internet  `0    S (  HT H  "Ht  c 6AARTBANNA"@h  C *AArthsepa" H  Z   ȜȜ?"p  T Click to edit Master title style! !  0kF "   W#Click to edit Master subtitle style$ $  6pA "D  0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  F* b  60F "  0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  F* b   6 "O 0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  B* `T  <d޽h ? MMM̙f  | t l (  l l s *' P   0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  P*   l s *P=    0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  R*  d l c $ ?   l s *0R  @  RClick to edit Master text styles Second level Third level Fourth level Fifth level!     S l 0@o `P  0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  P*   l 0 `  0___PPT106___PPT9B___PPTMac11   hnamd` Arial&Monotype Typography  R*  H l 0޽h ? 3380___PPT10.'u8 @$( M  r  S @9p  r  S    H  0޽h ? 33___PPT10u.%u5ɩ+D=' = @B +}  P(  l  C  `0   l  C @j `  H  0޽h ? MMM̙f___PPT10u.&ul+D=' = @B +}  `( 0 l  C  `0   l  C г `  H  0޽h ? MMM̙f___PPT10u.'uIi+D=' = @B +}  p( p l  C @A `0   l  C `B `  H  0޽h ? MMM̙f___PPT10u.'u@5+D=' = @B +}  P(  Pl P C O `0   l P C P `  H P 0޽h ? MMM̙f___PPT10u.u?+D=' = @B +}  8( ? 9Z 8l 8 C X `0   l 8 C X `  H 8 0޽h ? MMM̙f___PPT10u.u$+D=' = @B +}  T( &`U@# Tl T C v ` 0   l T C ɶ `  H T 0޽h ? MMM̙f___PPT10u.uH+D=' = @B +}   (   l  C @]] `0   l  C `^] `  H  0޽h ? MMM̙f___PPT10u.(u}+D=' = @B +}  X( ?fFA>> Xl X C }[ `0   l X C ~[ `  H X 0޽h ? MMM̙f___PPT10u.uzQ+D=' = @B +}   ( f  l  C  _ `0   l  C  _ `  H  0޽h ? MMM̙f___PPT10u.(u{/+D=' = @B +}   $(  $l $ C  ` 0   l $ C  `  H $ 0޽h ? MMM̙f___PPT10u.)uFC+D=' = @B +}   ((  (l ( C  `0   l ( C  `  H ( 0޽h ? MMM̙f___PPT10u.)uL+D=' = @B +}   4(   4l 4 C R `0   l 4 C S `  H 4 0޽h ? MMM̙f___PPT10u.+u+D=' = @B +}   \(    \l \ C pH `0   l \ C J `  H \ 0޽h ? MMM̙f___PPT10u.uXm/+D=' = @B +}   `(  `l ` C p `0   l ` C p `  H ` 0޽h ? MMM̙f___PPT10u.uc+D=' = @B +}  0d(   dl d C w `0   l d C w `  H d 0޽h ? MMM̙f___PPT10u.u+D=' = @B +}  @,( N$A[& ,l , C 0ky `0   l , C Ply `  H , 0޽h ? MMM̙f___PPT10u.*u-+D=' = @B +  P0(    0l 0 C 0{ `0   l 0 C P { `  H 0 0޽h ? MMM̙f80___PPT10.+u:  `@( &`U@# @l @ C { `0   l @ C { `  H @ 0޽h ? MMM̙f80___PPT10.ut  h( Ksk hl h C  `0   l h C `ˆ `  H h 0޽h ? MMM̙f80___PPT10.!u |  p<(    <l < C `| `0   l < C | `  H < 0޽h ? MMM̙f80___PPT10.uLd  H( &`U@# Hl H C  } `0   l H C PF} `  H H 0޽h ? MMM̙f80___PPT10.u|  D( !A1 Dl D C =~ `0   l D C p>~ `  H D 0޽h ? MMM̙f80___PPT10.u7  0L(  Ll L C 0۽ `0   l L C  `  H L 0޽h ? MMM̙f80___PPT10.u@ tlp(   pR p 3 l   r p # l @    H p 0޽h ? 3380___PPT10.'u8@ tlt( p` tR t 3 l   r t # Pl @    H t 0޽h ? 3380___PPT10.'u8@ tlx( K xR x 3 l   r x # l @    H x 0޽h ? 3380___PPT10.'u8@ tl|(   |R | 3 l   r | # l @    H | 0޽h ? 3380___PPT10.'u8@ tl( 71D R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8 @ tl( \  R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8@ tl ( %K R  3 l   r  # @l @    H  0޽h ? 3380___PPT10.'u8@ tl0( %  R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8@ tl@( K R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8@ tlP(   R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8@ tl`( \ @ R  3 l   r  # `+l @    H  0޽h ? 3380___PPT10.'u8@ tlp( \ " R  3 l   r  # 8l @    H  0޽h ? 3380___PPT10.'u8 @ tl( %  R  3 l   r  # PFl @    H  0޽h ? 3380___PPT10.'u8@ tl( \ !/ R  3 l   r  # Tl @    H  0޽h ? 3380___PPT10.'u8@ tl( \ AB R  3 l   r  # Pal @    H  0޽h ? 3380___PPT10.'u8@ tl( \ D R  3 l   r  # nl @    H  0޽h ? 3380___PPT10.'u8@ tl( \ P\@e R  3 l   r  # {l @    H  0޽h ? 3380___PPT10.'u8 @ tl(  R  3 l   r  # 0l @    H  0޽h ? 3380___PPT10.'u8 @ tl( \ Pk^ R  3 l   r  # pl @    H  0޽h ? 3380___PPT10.'u8@ tl( \ ^ R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8 @ tl( \  R  3 l   r  # pl @    H  0޽h ? 3380___PPT10.'u8@ tl(  R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8@ tl ( \ д R  3 l   r  # @l @    H  0޽h ? 3380___PPT10.'u8@ tl0( \ 5P R  3 l   r  # l @    H  0޽h ? 3380___PPT10.'u8dxp^RЀ3ÿ lHbP  @AL G@;b `B&V'@7( /DocumentSummaryInformation8Current UserA Oh+'0DL hp  4 @ LX`' FIND experimental requirements David Clark@Macintosh HD:Users:ddc:2006-active:Talks:Future Internet.ppt David Clark14Microsoft PowerPoint@{z@@UԲGJPICTJ HH HH  hhP=E21&'NRRgkQKUF4  +!Fdb_QzN?@DPQn{Fdwa1AUSCI2Awz^RS>&]g.`%pJYÖc5454345545l|ļ;Ţh#$&%!)F>%$.6irriOEOk0+*&$# #"$ "  #30V,'C"#O!>`M.4#%+!#'#ID9( Sv`NU7" !  -S.HS/:-*SsĪ>! ! !'6Y\u9|bRXTRSV^QWplTRl`82D9;L&-%!;P]\@VfSPHDQLqnQr_bSA7ZYI.;NmtmZSH=bh6=CBXtœW?>==:99>EGsh!#.*)/3*"#0)bG|j~iF4Py:34)&# $!   $&(B06&#"#'!@Y<.<$&+#& CP?$#5E90E1 n##F*>GG(O_|02Dl_gor{eGKM@GXULHML@FiXx;( R_K1Hozh         B        # ##$$#$)$ ## !$#! `aS?G>8rwfzzpvF4=+##  !*)$%6R79_AaykeT[Yy{\{~SiMLW^QG5OtbHE`zzgYTRPmV3ya3>efFJMHGEHLOUnW!"*&17.0*4. j~}\lRh}w{gmu}s~{cB/>C8;9/)-+!"# ##6CC4,#()ES=9:C;..2DHpfTRqztp~~~rwjD=\}xw~mzrfzr_`I;IROSZ_Z]^eorj]bc^^`akc]yg`#]XRZ`L/64+3:KHENB6+,TaWJWZbXYYScli[WnddUUSUSSK+40$'+')'&#@%(#"!!!#"! ## #!%%#!CPI?5535BE?B?-'%-($03$$/GPKHHFC>7" # !#"%#"%"5/88,72IL>?#$+%"$"")(')+-524SODHQWqlYs{kro|jK[y{sej}fHWC*%./4>PTST^lyvfrxpvsp=m{maW[ZRSPSJ9C;.46KKGP7.21NIF;A7"'# ")+'$$,0+$##"%##3637OU\KCD=81>GTK90'#&*-,069JTSRPPKE;/"# !#$(1,)/!  ""&*'.0'+ +3-A7**<GHKO,$,& %&'!(\XW\\beHDB?W  !!"#%"" !"!" $  !!9           NRTRKOOg{|}yc[eiSaYX`hets{ªwtmu·q_xwpig}YWL((#-7EIQUces}yux}x}v{hWR[*WMPEIKDL?:IADIJO=/$%:AB;B:IPLMQSYUNHOVONJMOS\ZXUPF"#%(/169;:5.$#%'##!#"#!$6%#9?74DRQbOKM@9CKTTR<4)&().2=DHOPTPOUWN;;-!#!"##)0310:-  !,(    [ *$"$.+/@G+.0.&(#,,*""LLQMP[ZOL45 " !"#"!" !%'('"#&"!" ##!!$"" #""$'$%# "$!!       dQUYXTS`qôpx~jius{vȲ~~yƼjwssukxmHP2"&:?CKT_dnz{{w|zwyo[NOXPENGDHJHFFTNIHKKEG6")GI/3?MPOQ[NBEMSRSTKPSTUgb[hQP&#"(2>FHLMDC9*(-&#-&"&/&:H::JG9@RPRcSDLNX[T^_TG<410,18KSPMIMST\VL:BF"!!#"#$+1576?@   %     :     !#$.:9?24170/.*7/#&;:7;4CKFM31 !$$ ! # !"!%)*!".'"%&($%*)'&%$%'$##" "##&(#"$( #!      QaSba_USap~umr~Խq||z|~uKT=1=>MUY\gwuuhpw>{k`\VKQSIAF;:EFMSPSSNINKB<3&%:<14;CITRLGJFEV[MKRLJOTTY\b\WXW%"#"$-5FMMRTPC86-$"$22)/58FRA>RVTUXRN[ONY]_cad`YI@;566;@JSSQMGUabRNMNT4!#$##%283449D"   !      .  !!*-/5347527<1% "%)%&.0./0-<>#"#-$$#$!"&,$#01)*0%"##$$',),-0/050./-1/*'#!$*-*,046!#&*"       JWkve_drw¾y|{{HME-9;DQ_]`hi~~pnuzfTSRITRJI@2:GHOUZ^]PLPQH:259:1/8GSTYXTX^UPH?<:=6=DGQTN@.!##%#&*     $MXjvorżج~zwWCS0#8?BNUap{vljhwjsrUSOVfdTNG=DFCLUZcqsjZSSOOPQVXYage^\cbVMCIFLOMJh@NA=@KBEMFGG"$*$'4?HA87AHAGMK8&)94**,6:8;C?9=?1/779;?@;AHIXSRR@IO@L?G^]GTOIC0DED?C@1++%1#5 ##""# &-/$" #! " #&(,20-+1/560/56330*35874("##,6-58:90")/*%'  ##(:-25#VipnſËxxg8KC& 0:HNWYo}jglmzvw~{}xVPQ]|r`UQGLMIKPZl|t]RRTVTOWcdhjiqegh_KNPEDBBK?JLeJPUOQKF""%29:8:8357CC>=:;3696/))/31/./+09579@:72867FOSTY\`cl[3)$#+16=@9:;7  !   * !)3MTXbtwwoqt~oPNQgveTSNPRU]Zdoyyn`VZ\][Q`ikggitmqmhN=LYUFFTYidhog\q{pre]#"(3<7579889;EB97;>;:<65)(')06;@?B>=??=8785>EGGJQSI?FQTSSTdaXPHEHJPRRZggfhi[E0$"'2./66:9:(   "    ) #%5@DJECLLIJF4.)$-#-=:E9?:$<&!"#"!#"!" #&*.//,)+*+(*+#,.-331//./121/,*),05:85312+#!%#%   ؁k½ε|wzrYbN$!!09?ILWmx~rxy}vQHOkta]RSQVakxuusl_]ahpqmj_nvtukowwyjc\SLUWYelmjmxsbpyio#)365468@@97>FB:89@@::3)(''*9AIPQPPD??<874=GPPS-TNBENRQRVd\WLDKMQSSTdjhhjldQ4'"#*/7=;68>9# %!   ) '1?JJTFGOQSGD87.(*+HPORIP90UC?!$"# ! "#$*./2--+,+()()*+//2330./12320.,02379855453'!##!#% ؁$x¸лxy^^_0!%6@>HNS`{zvzt~OFQipf[Ta[UamtuusadbV`sxspmr||~|{{ugcls`QEQ\a9WSovnanitqhp&6342469BG@6=BA=9=@<5,'(*++5@KNRV^UNB>;757>LVVS=RHCGIIHPWeYSKDGHHPXZefihjmf[;-$&'.5E=45;; "#   & "15>PRSPOSSOF@:2%!;LF-+CP@"**40#-!#1!!"# ! "#&(*+//.-*()+)&#&&+./133/.392/27587967735,## !"   ľ}z^WxD(9?>KRYbzd[bmfIK\qfXahko^]lrjhspmmjfovyuwz|wunluoYQTSZWLQky}jitmbc[58947:8:?B?89>?>',AJ1#(5QU)+#&)0##$"!N ""$&),..+-/1..,*+**+*,-1363244322878774/00/17?<7620%##"#   tıroosszeMG+GfzY]otvqstYqðҺogsxsbwV%)47?KOMYuvwh`^UWXWbdM@??EOZRSeih`ZdlhgmmomacddZ_ep|yotnhokaUSfN>JWVURV^jmubbeYLKL:=@@84985)%)''(/:BCBDJNMG@;3)&"&5G;32=9>@AJNPUVPL<<>>0@IX_`]^dcZQ;5:/Q]V]mlfVK4)#'+29FFDJH4$,"!%%$')6@ ?:5+$$ 8 (59@FPROQQMRNHE4?'!1DIDEILJPQ26;BFJL  =!#%+.)!! !"()..-+*+..350/+-(.1369;<::5/&"%)6@D:76/' #$"#%  4@ ?:5("  $ &  4EJLKQNN.#&!  !$ #$1LA<1 $##"$+//-*,#0G[^c[VNKKLX[`cjP<:=:5003;?Q}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ڔ066]];@DGGggGIKrKLL#*0E ~0X66{6;@DGGGIKKLLY_L#*0'- '0632;@DGG&&GIKE6KLL=L#*00w~~e0R6]];@DGGSSGIKjKLL#*0EyhE0666ƹ_k̊;@DGGbGIKYKLLY_L#*0'*0 '0"663$0/6;@DGG=?GIK;(KLL=L/#*00006]]66;`;>c@@DvDgUIIVjLL#*0E00̙66Ƭ66;;>̭@Hɟ@V`DSD̚Gɕ̫IP̕I̫KK̾YY_L#*0' 0 '0/63-66;!;>:0@86@ 5 DD+&D@+/C/9IC ?I"':G.F*L=L6#*00O0yF]66;U`;?r@@l@[PDPpyhś`KLL#*0E̽00X{6Ƭ66;|;?aPɟ@̓̂P̧Lf̄DDĉGUrWPsso_t_YY̋Y_L#*0'#0 '0 - 63,66;!!;>:77@$ '<46@ +3>@ -@(8((C+B B)I-I5HF, K6%J#KAG=L6#*00޽½A06]66;Uj;?Y@@@DPGggGGIKKKrKLL#*0E̠006{6Ƭ66;|;?̬@@̦@PSDq̱̉SɅGGɀGGP]Į̧KK_YKK̆Y_L#*0' )0 '0 ,5! 3,66;!!;>"'@@5@4DD-&&&@CG((CCG0G<E"'"<;KCK;*KH;.G=L2#*000066]66;`;?Y@M@DPİ~GggGGIKK׬KjLL#*0E;006Q6Ƭ66;;?Ɗ@@̴@PSDq̉̌LɅGGɅGGW]ɱKK__K̵̆Y_L#*0' 0 '0 ,6, 3,66;$;>-0,5@@0@4DD-&4313@BG((CBG9G;E0772C;KGK;*K#,6H=L1#*000066p]yy;?@@u@DdDvGggGGIhIKK׬KKVLL#*0EM00666ƹ_k̅;?T@@~̏P̐Dq{DDɅGGɅGGW]I]̎KK__oKo̖Y_L#*0' 0 '0 ,66 3%1/7;>;<1@ 094@4C@DB@DGIEHIKJJKHHKLP#*06;@DGIKL#*06;@WZDGIKL#*06;@2DGIKLN#*06;@lPDGIKL#*06;@dDGIKL#*06;@?%$0=@6 @9D9D+-8 C!5G-#/HI.)IE#C.G(.L*06;?`ڣMuDDenGgpqIIUrVKL*06;?|̚|̥@HHWDSZS̭hUɠ̅|̱]Ƨi̅II_KY}RɆL*06;?",0.@ >)?@ 6D7D&&7-8"F==)@/5@>7CE 0E.F*HA>GCGL*06;?`U;@@DDeeDv^GIhKKrKWL*06;>|;T̷yy@@WDSZSSD̮ĀGUmI̱WɂKjKR̵L*06;>!!;0$:>@7@7D7D&&7DD&&+((G;FI9)))EGK6*KFB ,CL*06;?``;ڪ@@㤞DDeeDvgGIhűKKrK`L*06;?|;L̚i~@@WDLZSZDƍGUUI̎PɆKYKKd̨L*06;?!!;0,00,;>@9@7D8D&&7DD&750;((G9GI;)8757EFK;(KGKC3?L*06;?``;@lcueeeDGggGIUIyKKr`KL*06;?|;D̞@~d@WDZSZD\GGU\Is}IIɆK_KjoKL*06;?!!;32@' ?*@'@+D$D&&7DDGB7((G9GI;0EI'EFK;(K:8JK4@@C@DGCDGIFEIKHKKGHLN*06;@@DGIKL*06;@PW@DGIKL*06;@2@@DGIKLM*06;@cY@DGIKL*06;@^@@DGIKL*06;@?<@@DGIKL:06;@DGIKL06;@DGIKL06;@DGIKL:06;@DGIKL06;@DGIKL06;@DGIKL:06;@DGIKL06;@DGIKL06;@DGIKL:06;@DGIKL06;@DGIKL06;@DGIKL:06;@DGIKL06;@DGIKL06;@DGIKL0Xnb`WJOD@<0?OK=72/00=>NcbQBCVS[`X`@D;Zb^d^SWYMLBP]cbcd@GC`^bbV`LIVhmZFG8896BLM>9<9;53544A6<=8752331(56:@B9D:7:79@G<=C97508?;;?87;;:=?CEM>GJLNL0 ) $'%#$"!"%&%$+*)+-*)*+**)(--/-.//0/--,-../.--12234.-/../. /101462211/L:06;@DGIKL06;@DGIKL06;@DGIKL46;@DGIKL6;@DGIKL6;@DGIKL46;@DGIKL6;@DGIKL6;@DGIKL46;@DGIKL6;@DGIKL6;@DGIKL46;@DGIKL6;@DGIKL6;@DGIKLm6;@DGGGGGIIIIKL6;@DGGGGGIIIIKL6;@DGGGGGIIIIKL6;@@DDGGGGGGIIIIIKKL6;@@DDGGGGGGIIIIIKKL6;@@DDGGGGGGIIIIIKKL6;@@DDDDGGGGIIKKL6;@@DDDDGGGGIIKKL6;@@DDDDGGGGIIKKL6;@@DD DDDGGGGGGIIIKKKKL6;@@DD DDDGGGGGGIIIKKKKL6;@@DD DDDGGGGGGIIIKKKKL;@@DDDDGGGGGGIIKKKL;@@DDDDGGGGGGIIKKKL;@@DDDDGGGGGGIIKKKL;@DDDDGGGIIKKL;@DDDDGGGIIKKL;@DDDDGGGIIKKL.;@DGIKL;@DGIKL;@DGIKL.;@DGIKL;@DGIKL;@DGIKL.;@DGIKL;@DGIKL;@DGIKL.;@DGIKL;@DGIKL;@DGIKL.;@DGIKL;@DGIKL;@DGIKL.;@DGIKL;@DGIKL;@DGIKL.;@DGIKL;@DGIKL;@DGIKL(@DGIKL@DGIKL@DGIKL(@DGIKL@DGIKL@DGIKL(@DGIKL@DGIKL@DGIKL(@DGIKL@DGIKL@DGIKL(@DGIKL@DGIKL@DGIKL(@DGIKL@DGIKL@DGIKL(@DGIKL@DGIKL@DGIKL(@DGIKL@DGIKL@DGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL(DDGIKLDDGIKLDDGIKL"GGIKLGGIKLGGIKL"GGIKLGGIKLGGIKL"GGIKLGGIKLGGIKL"GGIKLGGIKLGGIKL"GGIKLGGIKLGGIKL"GGIKLGGIKLGGIKL"GGIKLGGIKLGGIKL"GGIKLGGIKLGGIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLIIKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLKKLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL ՜.+,0P     On-screen Show David Clark ArialMS Pゴシック Wingdingsヒラギノ角ゴ ProN W3Future InternetFIND experimental requirementsFINDThe original Science PlanGeneral requirementsBackground commentaryThe excitement is at the edge#The excitement is at higher layersNew protocol stacksLower level research?Packets+New functions “in” or “on” the netEmulating a real networkOptics in the coreOne alternativePicking compromisesAnother way of saying this ManagementA general challenge for GENIScale and speed Real scaleSome specific requirementsInstrumentationNon-requirementsThe experimental landscape  Fonts UsedDesign Template Slide Titles#_ @DDavid ClarkDavid Clark      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLM 00DArial ProN W3ڄnlD0D-3 00000oN W3ڄnlD0 DWingdingsN W3ڄnlD00D0000҉0 ProN W3ڄnlD0 ` .  @n?" dd@  @@`` z43        !"#$%&'()*+,-./0123?b$ ە/+h;K_b$XO w/\c b$`B;;˃ i;S @8y]uʚ;2Nʚ;g4UdUdCмbppp@ <4!d!d0ׄgʚ;<4dddd0ׄgʚ; :\___PPT10  H P (P HH`   0 8(P 0 ( 0H 8(0N ___PPT90 ( @@  P    ` pP 0` P ` pP`h___PPT2001D<4XI___PPTMac11I@f   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography B0   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  0   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  d   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  h   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography $   hnamd` Arial&Monotype Typography  h   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography D   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography 8   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography   ,   hnamd` Arial&Monotype Typography $   hnamd` Arial&Monotype Typography  h   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography ,   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  8   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography  $   hnamd` Arial&Monotype Typography   8   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography   ,   hnamd` Arial&Monotype Typography  4   hnamd` Arial&Monotype Typography     hnamd` Arial&Monotype Typography H   hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography    hnamd` Arial&Monotype Typography ? %HFIND experimental requirementsDavid D. ClarkFIND%Future Internet Design (FIND) is an NSF program (now folded in to NetSE) to envision the Internet of 15 years from now. FIND was one of the early drivers for the development of GENI. FIND research spans the layers from technology through classic Internet architecture to application design. 0&y@l >B0>mThe original Science PlanBAs part of the original work on GENI, we prepared a Science Plan (also called the Research Plan), which listed various requirements. Those requirements have not changed. The GENI plan, as then proposed, (and now?) did not fully meet those requirements. Suggests a need to revisit some assumptions and design approaches.>ZCZ&TC Pg$RCGeneral requirementsOA real, distributed network. Not a bunch of routers in a rack. As wide a reach as possible. Reach to the edge. Allow experimental edge equipment direct connection to the GENI-based experiments. Access to real users. Creates a tension with the desire for realistic lower layer technology, e.g. optical layer. Long-running experiments. ?S\?S  \wBackground commentaryThere are a wide range of experiments (with a wide range of requirements) that might be posed for GENI. It is not realistic to imagine that we can build a single fully general facility that can support all of these experiments at reasonable cost. Implies a need to be creative, make clever choices. Make some compromises. The MREFC process distorted the translation of requirements into facilities E.g. a single unified facility. All of these assumptions should be reconsidered. One or many? Is GENI one thing? A better set of outcomes?ZKZLZ Z1Z:ZhKL   1   PIQB - The excitement is at the edgeWhile (some of us) want to mess with the core Security, management, etc. The real action is at the edge. New devices (mobile, embedded) New networks (wireless of all sorts) Not all parts of the network will look like ethernet! Cars and other mobile networks. To emulate the future, need all this in the experiment. Not so clear how to virtualize. Does this matter? Remember the goal of reaching real users. /ZZ ZDZ6Z!Z8Z]Z/   %6!82  +$$- S  ="The excitement is at higher layersDesign patterns for applications. Highly distributed, clouds, etc. New support mechanisms Identity frameworks, location frameworks, etc. Important to ask: to what extent can we explore these on the existing Internet?b""/P""/ P New protocol stacksResearchers want to try new protocols at the network, transport (and higher) layers. New means of authentication. New mechanisms to deal with soft state in the network. & Implies the need to replace the protocol stack in the end node. Facilitating this should be part of GENI effort. Mobile devices, not just PCs. Remember, we want real users. A real experimental tension here& VZVZAZZV7  A1  "$$,TkLower level research?Two sorts of reasons. We need to do a better job supporting apps. Security, availability, management, economics, etc. (It is not the data plane, except indirectly& ) What happens there will influence the design of applications. The research is not independent. We don t have the lower levels right. Management, operations, etc. Some of this is perhaps more independent.Z,ZZ!Z&ZHZ,4. >!&H Pb* PacketsWe want to try out packet formats that are not IPv4 and IPv6. Congestion and its control. Novel addressing modes New mechanisms for security. New concepts in network management. New schemes for tracking payment. & This capability must reach all parts of the experiment. >:>  $"9   DNew functions  in or  on the netNot all boxes that are topologically  in the network are routers. Security enforcement devices. Encryption devices. Packet inspection devices. Application support devices. & Implies different node requirements. More storage, processing, etc. Very high performance. Network devices today are highly purpose-tuned. How can we provide generality for a range of experiments with different topological requirements? Processing nodes should be in the net, not just at the edges.CZlZ%ZZC  %g  b$$=((,,b$~;F Emulating a real networkThe core of a big ISP today does not forward packets, but is built of flows that carry aggregates of packets. Optical lambdas, MPLS circuits, etc. Not all parts of the network will look like ethernet! Cost and complexity is a major driver in real nets. A future architecture should do a better job of linking the management of these layers. How should these capabilities be made available to experimenters? Should this be a GENI goal?jnZ\ZZn\5X  ]b$3$, Optics in the coreThe original proposal for the GENI platform had rather sophisticated optical components in the core. E.g. ROADMs. This had major cost implications. This had major  virtualization implications. How do you virtualize a ROADM, since it is not a packet device? Not well done in original proposal. The goal was to at least emulate how real networks look today. eZ ZPZdZ@Ze P@  #@P/:2* One alternative`Let the folks who want to play with real optics build their own environment. Smaller scale? Build the large scale testbed out of packets (IP, ethernet: does it matter?) and tunnel our  new packets inside them. Is this a better idea? (It has limitations that should be recognized, as well as benefits.) bNZZZZNw[ bU=>Picking compromisesIf we fold optics in& More realistic (but for what class of experiments?) QoS, non-packet end-to-end services. Intrinsic availability, security, management, etc. Cross-layer protocol designs. If we use packets and tunnel& Much easier to achieve scale and reach real users. Lower layer  technology complexity will have to be simulated. Is this an issue?4v4%3  3P  $$JAnother way of saying thisThe independence of the parts of the Internet (e.g. apps from link technology) is as a result of the  hourglass design, the end-to-end design, etc. Assuming that experiments can tunnel over packets risks baking in today s hour-glass architecture in tomorrow s experiments. My opinion: the risk of guiding research toward a presumption of the hourglass has to be mitigated in some way.6p}pp  ManagementOAs the previous slide tried to emphasize, a lot of future Internet research is about management, not the data plane. So GENI is not just virtualized data planes, but virtualized network management schemes. Fault diagnosis. Virtualization messes this up. Set up and tear down of circuits. Want to mimic real operators, not just users.jZZZPZvZ ".bE0  A general challenge for GENIMany of the proposed ideas for a future Internet stress management, security, etc. It may be less clear how to virtualize the experimental infrastructure to allow these to be demonstrated than to virtualize the data plane. For example, to demonstrate improved availability, the GENI platform would ideally mimic the baseline failure modes of the eventual technology. What is the best set of compromises? Again: one GENI facility or many?b%"S%  "t0! "'  Scale and speedSome experiments stress scale 100 s of routers, 100 s of fixed end-nodes, N00 s of mobile devices, and rich connectivity. Others mention  thousands of interconnected devices . Speed was not an issue for this experiment. These are miniscule experiments! d!\5 ," Real scale10K-100K AS regions. Routers with 500 10 Gb ports. 100 striped lambdas& Millions of multi-homed end-nodes. Highly heterogeneous environment. How are we going to try (and validate) ideas at these scales? 3ZZEZZ?Z# "= b! 2[ Some specific requirementscRich connectivity (to experiment with novel routing.) Need for multiple regions (emulate different operators). High bandwidth paths between slices? Heavy-duty isolation among slices. Detecting physical location. All devices should be able to do this. Universal crypto capability. Allow experiments in virtualized architectures. Recursive virtualization?oZ%ZAZ(ZMZZ69%A  (MPn JInstrumentationNIn some respects, a confused discussion. Clearly, we need to gather data on how experiments are going. But is this done in the infrastructure, or in the slice? A future Internet must include native capabilities for instrumentation. Is  data something generic that is shared? Wishful thinking?Ng>)>:I >P(A4G>Non-requirementsSatellite Not a sufficiently distinct challenge. Residential access networks Again, a challenge, but not sufficiently distinctive. If have wireless nets and high-bandwidth links to end-nodes, good enough. x '6J '6 JuXThe experimental landscapeHow many experiment? PlanetLab has demonstrated that there can be 1000 s of experiments. But how many folks want to try out a new Internet? Perhaps we need different sorts of tools for experimental deployment at different  layers . Again, one GENI or many?XD4\ , lu/ !"#$%&'() * + , - ./01234567rA'A7