Ilija Vukotic Data size and IO performance report ATLAS Software & Computing Workshop
Data size –MC infoIOperformance/MCsizes.htmlhttp://athena-infoioperformance.web.cern.ch/athena- infoIOperformance/MCsizes.html –Streams infoioperformance.web.cern.ch/athena- infoIOperformance/StreamsSizes.htmlhttp://athena- infoioperformance.web.cern.ch/athena- infoIOperformance/StreamsSizes.html –Categories infoioperformance.web.cern.ch/athena- infoIOperformance/T0streams.htmlhttp://athena- infoioperformance.web.cern.ch/athena- infoIOperformance/T0streams.html IO performance Possibilities for improvement 05/03/2016Ilija Vukotic2
MC event size 05/03/2016Ilija Vukotic3
MC event size 05/03/2016Ilija Vukotic4 Note: trigger size is not realistic as used trigger menu was not realistic. Problem rectified with addition of new aliases: Physics_default -> Physics_pp_v1 MC_loose_default -> MC_pp_v1_loose_mc_prescale MC_tight_default -> MC_pp_v1_tight_Mc_prescale (see bug report #74712 ). Size AODSize ESD RecoTrf %* % stdSim % % fullSim % % physSim % % *Compared to 22 Aug. 2010
Streams sizes 05/03/2016Ilija Vukotic5 Only runs with 5k+ events and good LBs Shows: AODs,ESDs and all DESDs Events/total size/ev. size
Streams sizes 05/03/2016Ilija Vukotic6 StreamSize[GB] ESD AOD92319 DESDM_EGAMMA55288 DESD_SGLEL44191 DESD_SGLMU39375 DESDM_TRACK26548 DESD_MBIAS16841 DESD_CALJET12370 DESDM_MET5108 DESD_PHOJET4281 DESD_SGLMU DESD_SGLEL DESDM_EGAMMA AOD ESD
05/03/2016Ilija Vukotic7 Categories
05/03/2016Ilija Vukotic8 EgammaL1CaloMuonsJetTauETMiss Details and list of collections in categories can be found at:
05/03/2016Ilija Vukotic9 IO performance Slow Bad CF KEEP IN MIND In real data TrackCollections are even larger !
05/03/2016Ilija Vukotic10 IO performance Do we really need all of these? No way to say if object is ever used. What is so terrible about having a name attached to each object we store?
Possibilities for improvement TrackCollection/TrackParticle container –Maybe 50kb in size and factor 2 in time –Few MB in memory CaloShowerContainer –Probably can go down in size Calo(Topo)TowerContainer –Gets recreated at read time –Must be possible to optimize further TRT_Drift_Circle_Container –Must be faster and smaller 05/03/2016Ilija Vukotic11 My estimate: In total between 50 and 200kB/ev can be shaved in converters Should be possible >100ms/ev (currently at 480 ms/ev) Other options: Object removal. Cuts tuning