ࡱ> ;=: bjbj 4$]6WWWWWkkk8 kQh$[bWWW WWpX[?k!0Q$dW4Q : GMOC Coordination Meeting Stephen Schwab, SPARTA (GSAT Project) Jon-Paul Herron, Indiana University, (GMOC Project) July 28, 2010 The meeting took place by phone. Discussions focused on what issues the GMOC is currently looking at or forecasting on the horizon, and where security input or review would be helpful. Jon-Paul identified the CONOPS draft as being worthwhile to review and provide comments on those comments are included below. Also, Steve Schwab asked what sorts of data the GMOC was receiving or expected to receive in the near-term. This data appears to be slowly growing, and may include: --ProtoGENI data (possibly the Internet2 HP openflow switch static topology configurations) --PlanetLab COMON data --GpENI - data from local GpENI clusters Jon-Paul also provided a local deployment diagram of the Enterprise GENI openflow switches in the Indiana University environment (Jon Pauls IT organization) as an example. -CONOPS document -- early draft out now the following comments were made to the posted GENI Concept of Operations document, GENI_Concept_of_Operations-final.pdf. (Need to add revision numbers to next version.) Comments -- pg. 1 -- "Third, GENI could choose to leave operations of GENI components and sstems to individual GENI aggregates and control frameworks, but provide an integrated suite of services to coordinate operations, and provide the unified interface for GENI operations to the GENI users and stakeholders" -- --- You should comment on how this interacts with the 'clearinghouse' model, where users already expect to see all aggregate/component resources from a single integrated interface -- is what you are suggesting similar to a clearinghouse, but for the operators as well as the users? pg.1 -- "This will trigger the requirement for a unified contact for experimenters to report trouble, as well as the system to connect those GENI experimenters with the correct GENI federate operators..." --- It seems more likely that users will be getting information from their local control framework or the interface that directly lets them create slices. So the situation may be exactly reversed. A failure will be detected locally by the user and reported to the control framework. It would be useful if other users or aggregate managers could browse 'notifications' that summarize these sorts of local failures, to more quickly diagnose what is wrong. This might also avoid flooding a control framework or aggregate with identical inquiries when something does go wrong and all users, local and remote, are impacted. pg. 2 -- list of stakeholders -- probably need to call out explicitly the campus IT operations staff who are not explicitly GENI researchers, but must support the deployment of GENI aggregates and the set of GENI experimenters (researchers) on their campus. pg. 2 -- experimenters -- also care about the availability of resources (components and network connectivity) that they need to allocate to create their slice, e.g. they care about what is available prior to setting up their slice, and then they care about the 'uptime' of resources in their slice. pg. 3 -- common GENI infrastructure providers -- it is clear that Internet2 and NLR should be able to get any information needed about other GENI elements, since they are the backbone. But what about 'similar requirements' for production R&E and Internet networks that interconnect with GENI? Who is going to decide if a "peering" exists between GENI and an "other" network, and what that other network should be able to see? pg. 5 -- one variation on the models worth mentioning is that of letting a GENI project voluntarily outsource some or even most of their operations to an entity like the GMOC. Some projects may want to contribute resources to GENI, but may not have the size or scale to provide an operational POC. A small institution with a single faculty member and just a couple graduate students might fit this description. They would outsource the front-line operations to GMOC or a control framework, but provide a local point-of-contact on a less time-intensive basis. This might be a reasonable trade-off for some projects to consider. pg. 5 -- Meta-operations -- it still seems that the control frameworks will provide the first-line of response for trouble tickets and diagnosis. However, once two or more control frameworks can directly allocate resources from a single aggregate manager, it may become harder to coordiante 'who is doing what' to the aggregate -- so there may be a natural role for Meta-operations if that scenario evolves to become common. pg. 7 -- Active Experiments -- it seems that experimenters have a home 'control framework' -- and that most notifications/interactions would take place through that interface for the near-to-medium term. pg. 8 -- Perhaps "end-to-end stitching" of slices across multiple aggregates should be examined as a motivating use case in the next draft.     Page  PAGE \* MERGEFORMAT 1 \]^`acdfgijopʾʈʈhs hHhs hHhs OJQJ"hHhH5OJQJmHnHu hHh" jhHh" UhHhs 5OJQJhs 5OJPJQJ\hN&ljhN&lUhBRhbxhbxhbx5hs hbx5CJaJ@t c  ?\]gdbxgdbx $a$gds ]_`bcefhijjokd$$Ifl4Fi'$$ t 6 6   OO 44 la $& #$/If $& #$/If dgds sqoojgdbxokd$$Ifl4Fi'$  t 6 6 OO   44 la$$& #$/Ifa$ $& #$/If21h:pBR/ =!"#$% $$If!vh555#v#v#v:V l4 t 6 6+,5559/ O/ O$$If!vh555#v#v#v:V l4 t 6 6+555/ O/ Oj 666666666vvvvvvvvv666666>6666666666666666666666666666666666666666666666666hH6666666666666666666666666666666666666666666666666666666666666666662 0@P`p2( 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p8XV~ OJPJQJ_HmH nH sH tH J`J BRNormal dCJ_HaJmH sH tH DA D Default Paragraph FontRi@R 0 Table Normal4 l4a (k ( 0No List >@> s 0HeaderdH$.. s 0 Header Char> @> s 0FooterdH$.!. s 0 Footer CharL`2L s  No SpacingCJPJ_HaJmH sH tH R/AR s No Spacing CharCJPJ_HaJmH sH tH PK![Content_Types].xmlj0Eжr(΢Iw},-j4 wP-t#bΙ{UTU^hd}㨫)*1P' ^W0)T9<l#$yi};~@(Hu* Dנz/0ǰ $ X3aZ,D0j~3߶b~i>3\`?/[G\!-Rk.sԻ..a濭?PK!֧6 _rels/.relsj0 }Q%v/C/}(h"O = C?hv=Ʌ%[xp{۵_Pѣ<1H0ORBdJE4b$q_6LR7`0̞O,En7Lib/SeеPK!kytheme/theme/themeManager.xml M @}w7c(EbˮCAǠҟ7՛K Y, e.|,H,lxɴIsQ}#Ր ֵ+!,^$j=GW)E+& 8PK!Ptheme/theme/theme1.xmlYOo6w toc'vuر-MniP@I}úama[إ4:lЯGRX^6؊>$ !)O^rC$y@/yH*񄴽)޵߻UDb`}"qۋJחX^)I`nEp)liV[]1M<OP6r=zgbIguSebORD۫qu gZo~ٺlAplxpT0+[}`jzAV2Fi@qv֬5\|ʜ̭NleXdsjcs7f W+Ն7`g ȘJj|h(KD- dXiJ؇(x$( :;˹! I_TS 1?E??ZBΪmU/?~xY'y5g&΋/ɋ>GMGeD3Vq%'#q$8K)fw9:ĵ x}rxwr:\TZaG*y8IjbRc|XŻǿI u3KGnD1NIBs RuK>V.EL+M2#'fi ~V vl{u8zH *:(W☕ ~JTe\O*tHGHY}KNP*ݾ˦TѼ9/#A7qZ$*c?qUnwN%Oi4 =3ڗP 1Pm \\9Mؓ2aD];Yt\[x]}Wr|]g- eW )6-rCSj id DЇAΜIqbJ#x꺃 6k#ASh&ʌt(Q%p%m&]caSl=X\P1Mh9MVdDAaVB[݈fJíP|8 քAV^f Hn- "d>znNJ ة>b&2vKyϼD:,AGm\nziÙ.uχYC6OMf3or$5NHT[XF64T,ќM0E)`#5XY`פ;%1U٥m;R>QD DcpU'&LE/pm%]8firS4d 7y\`JnίI R3U~7+׸#m qBiDi*L69mY&iHE=(K&N!V.KeLDĕ{D vEꦚdeNƟe(MN9ߜR6&3(a/DUz<{ˊYȳV)9Z[4^n5!J?Q3eBoCM m<.vpIYfZY_p[=al-Y}Nc͙ŋ4vfavl'SA8|*u{-ߟ0%M07%<ҍPK! ѐ'theme/theme/_rels/themeManager.xml.relsM 0wooӺ&݈Э5 6?$Q ,.aic21h:qm@RN;d`o7gK(M&$R(.1r'JЊT8V"AȻHu}|$b{P8g/]QAsم(#L[PK-![Content_Types].xmlPK-!֧6 +_rels/.relsPK-!kytheme/theme/themeManager.xmlPK-!Ptheme/theme/theme1.xmlPK-! ѐ' theme/theme/_rels/themeManager.xml.relsPK] $  5558])+8!T  # AA@0(  B S  ?enJRt~]]__``bcefhiovISw z \]]__``bcefhi33333\]]__``bcefhio" s BRN&lHbx$]_@x@UnknownG* Times New Roman5Symbol3. * Arial7K@Cambria7.{ @CalibriA BCambria Math"1hz #z #!0TT2HP  $Pbx2!xxStephen SchwabStephen SchwabOh+'0x  4 @ LX`hpStephen SchwabNormalStephen Schwab2Microsoft Office Word@@H?@H?z՜.+,0 hp|  # T  Title  !"#$%&'()+,-./013456789<Root Entry F@k[?>Data 1TableWordDocument4$SummaryInformation(*DocumentSummaryInformation82CompObjy  F'Microsoft Office Word 97-2003 Document MSWordDocWord.Document.89q