ࡱ> &RdO)0vOq@PowerPoint Document(*SummaryInformation(DocumentSummaryInformation8 ' !"#$%P( / 0LDTimesXC||XPsXs& ` .  @n?" dd@  @@`` P c $@T ʚ;2Nʚ;g4MdMds{Tm ppp@ <4!d!d`gʚ;<4dddd`gʚ; <4BdBd`ʚ;ph___PPT2001D<4X? %Infrastructure & Methodology(The cool group) Problem spaceWhat is  self-management anyway? Defined broadly - anything that does not require (or reduces) human interaction after configuration or indication of intention Insight: service level differs from device level Consequence: Human intention specified at service level Translated to device level Control loop handles issues when intention is not met t"ZZiZQZZ"iQYet more problem spaceTwo separate infrastructures are required One that accommodates the new self-management tools and methods (the real system) One for measurement & evaluation testing (the control plane) Models (abstractions) for configuration and workload are required for simulation and emulation The evaluation environment must provide level of detail appropriate to self-management system/method being considered\*ZZ_ZvZ*_v  Requirements Managed require extended capabilities Ability to express formal specification of intent Extended UPNP? Ability to express capability Ability to express current status Put into a model and a basis for control action v'Z2ZZAZ3Z'2A3  $0  Next stepsHow do we make progress? Establish a baseline for self-management Establish metrics and measurement methods/tools for evaluating self-management Establish evaluation infrastructures (simulation, emulation, in-situ) 6Research opportunitiesNetwork configuration (service level) specification Optimize performance (sla s) for minimum cost Best practices for translating to service policies given device capabilities Understand device/protocol interactions Define and collect metrics for improvement Down time/failures Volume of help desk complaints Human involvement Simplicity(?)v4Z.ZMZSZRZ4.MS  RJOpportunities continuedRBuild  Controllers for self-management components Define testbeds (simulation, emulation, in situ) Instrumentation to evaluate and validate Test cases Simulation models&dFdF:h  ` 33` Sf3f` 33g` f` www3PP` ZXdbmo` \ғq>9y`Ӣ` 3f3ff` 3f3FKf` hk]wwwfܹ` ff>>\`Y{ff` R>&- {p_/̴>?" dd@,|?" dd@   " @ ` n?" dd@   @@``PR    @ ` ` p>>  (    6~ P  T Click to edit Master title style! !  0~   RClick to edit Master text styles Second level Third level Fourth level Fifth level!     S  0~ ``  P*   0~ `   R*   0~ `   R* H  0޽h ? 33 $Blank Presentation  (  l  C p  l  C  `    H  0޽h ? 33  0 ( p*  l  C 0P   l  C   H  0޽h ? 33  `( | l  C +P   l  C .  H  0޽h ? 33   (   l  C 0P   l  C   H  0޽h ? 33  p( *xr l  C P   l  C   H  0޽h ? 33  @( h8 l  C P   l  C   H  0޽h ? 33  P(  ( l  C PP   l  C PQ  H  0޽h ? 33dxp^RЀ3ÿ lHbP  @AL G@;b `B&V