Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Multi-s ale Real-time Grid Monito ring with Job Stream Mining Xiangliang Zhang , Mi hele Sebag, Ce ile Germain-Renaud T A O − INRIA CNRS Universit� de P a ris-Sud, F-91405 Orsa y Cedex, F ran e 21 Ma y 2009 Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Contents 1 Monito ring system: Grid adapted StrAP 2 Streaming Jobs 3 Monito ring Outputs Monito ring on sho rt-time s ale Clustering Qualit y Monito ring on medium-time s ale Monito ring on la rge-time s ale Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Contents 1 Monito ring system: Grid adapted StrAP 2 Streaming Jobs 3 Monito ring Outputs Monito ring on sho rt-time s ale Clustering Qualit y Monito ring on medium-time s ale Monito ring on la rge-time s ale Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Multi-s ale Realtime Grid Monito ring System Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Multi-s ale Realtime Grid Monito ring System Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining 100 Percentage of jobs assigned (%) 8 exemplar shown 18 as a job vector 24 80 30 595 139 60 10 7 7 47 0 13 54 40 0 129 14 0 24 0 0 9728 0 20 0 19190 Outliers 0 1 2 3 4 5 Clusters
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Multi-s ale Realtime Grid Monito ring System Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining distirbution of jobs like [7 0 0 0 0 0] distirbution of jobs like [0 0 0 0 0 0] 90 30 80 25 percentage of jobs (%) percentage of jobs (%) 70 20 60 50 15 40 30 10 20 5 10 0 0 0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160 days days
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Multi-s ale Realtime Grid Monito ring System A�nit y Propagation (AP) A lustering metho d: group simi l a r p oints together StrAP (Streaming AP) Online Clustering streaming data based on AP Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Why AP ?? A�nit y Propagation (AP) A lustering metho d Group simi l a r p oints together Converge b y Iterations of Message passing � > mo re stable results No need of K (the numb er of lusters) � > less p rio r kno wledge A real p oint as an exempla r to rep resent a luster � > avoid meaningless averaged enters Clustering b y P assing Messages Bet w een Data P oints. B.J. F rey , D. Due k. S ien e 2007 Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Ho w AP w o rks ?? Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Grid adapted StrAP Grid adapted StrAP (Streaming AP): Online lustering streaming jobs � > one-s an of the stream In remental up date of mo del � > k eep tra king the stream Dete ting distributi on hanges in stream � > abso rb new patterns Data streaming with A�nit y p ropagation. Xiangli ang Zhang, Cyril F urtlehner, Mi hele Sebag. ECML 2008. Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Stream lustering Mo del Reservoir ❡ ❡ ❡ ✐ ✐ ❡ ✐ ✐ ❡ ❡ ✐ ✐ ❡ ❢ ❥ ✐ ❥ Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Stream lustering Mo del Reservoir Do es x �t the urrent mo del ?? t if y es, up date the mo del otherwise, go to reservoir ❡ ❡ ❡ ✐ ✐ ❡ ✐ ✐ ❡ ❡ ✐ ✐ ❡ ❡ ❢ ❡ ❢ ❥ ❥ ✐ Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Stream lustering Mo del Reservoir Do es x �t the urrent mo del ?? t if y es, up date the mo del otherwise, go to reservoir ❡ ❡ ❡ ✐ ✐ ❡ ✐ ✐ ❡ ❡ ✐ ✐ ❡ ✐ ❡ ❢ ❥ ❥ ✐ ✐ ❥ ❥ Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Stream lustering Mo del Reservoir Do es x �t the urrent mo del ?? t if y es, up date the mo del otherwise, go to reservoir �❅ ❡ ❡ ❡ ✐ ✐ ❡ ✐ ✐ ❡ ❡ ✐ ✐ ❡ ✐ �❅ ❡ ❢ ❥ ❥ ✐ Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Stream lustering Mo del Reservoir Has the distributi on hanged ?? CHANGE TEST if y es, rebuilt the mo del �❅ ✐ � ❅ � ❅ �❅ ❡ ❡ ❡ ✐ ✐ ❡ ✐ ✐ ❡ ❡ ✐ ✐ ❡ ✐ ✐ ❡ ❡ otherwise, ontinue � ❅ � ❅ � ❅ ❢ ❥ ❡ ❥ ✐ Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Stream lustering Mo del Reservoir Has the distributi on hanged ?? CHANGE TEST if y es, rebuilt the mo del �❅ ✐ � ❅ � ❅ �❅ ❡ ❡ ❡ ✐ ✐ ❡ ✐ ✐ ❡ ❡ ✐ ✐ ❡ ✐ ❡ otherwise, ontinue ❥ �❅ ❢ ❥ ❡ ✐ Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Monito ri ng system: Grid adapted StrAP Streaming Jobs Monito ri ng Outputs Our Mo del Output e , the exempla r ( enter of luster) i n , size of luster i , average distan e of p oints to their exempla r i T , time stamp when the luster w as latterly visited P a rameters ǫ , threshold of ompa ring ea h p oint with mo del (set to a round value of Σ in the initial mo del) i ∆ , de a y windo w (de rease the w eight of old exempla rs) P age-Hinkley pa rameters ( hange dete tion) Σ Xiangli ang Zhang, Mi hele Sebag, Ce ile Germain-Renaud Grid Monito ri ng with Job Stream Mining
Recommend
More recommend