1. ILD Software for ILD detector simulation and optimization Akiya Miyamoto KEK 12-July-2010 DESY Computing Seminar

2. ILD Contentes ILC overview GLD software tools and optimization ILD Software tools  Mokka  Reconstruction GRID in KEK 12 July 2007 Akiya Miyamoto 2 DESY Computing Seminar

3. ILD International Linear Collider : ILC e + e - Collider Ecm: 0.2 ~0.5 TeV  1TeV ∫Ldt = 500 fb -1 in 4 years 2004: launched 2007: ILC RDR 2009: Detector LOI 2012: TDR/DBD 12 July 2007 Akiya Miyamoto 3 DESY Computing Seminar

4. ILD Higgs – Vacuum Higgs coupling proportional to mass in the SM Did Higgs break the symmetry ? Condensed in vacuum ? 12 July 2007 Akiya Miyamoto 4 DESY Computing Seminar

5. ILD Supersymmetry Fermion  Boson Do forces unify ? Do Quarks and Leptons unify ? Mass, Coupling Extending to Higher energy/Early Universe 12 July 2007 Akiya Miyamoto 5 DESY Computing Seminar

6. ILD ILC Reference Design Report Released 2007 summer: http://www.linearcollider.org/cms/?pid=1000437http://www.linearcollider.org/cms/?pid=1000437 Executive Summary Physics at the ILC AcceleratorDetectors RDR: 4 volumes/ 774 pages 12 July 2007 Akiya Miyamoto 6 DESY Computing Seminar

7. ILD Challenge of ILC experiments e + e - Collision:Well defined initial states and relatively clean final states.  Find faint new physics signals  Make precise tests of theory Many events contains top, W, Z  jets  Precise measurement of jet energy Calorimeter system inside the coil Highly segmented calorimeter for high resolution  Efficient b/c tagging crucial Thin material, strong B field, VTX very close to IP, pixel detectors  Higgs recoil measurements ( e + e -  Zh  llXh) require good  P/P  Hermetic detector down to very close to beam pipe ( ~ 10 mrad )  Detector should be shielded well against beam related backgrounds, low energy e + e - pairs,   hadrons, muons, 12 July 2007 Akiya Miyamoto 7 DESY Computing Seminar

8. ILD Detectors for ILC experiments Tracker: Charged particles  pt/pt : 1/10 of LHC Tracker: Charged particles  pt/pt : 1/10 of LHC Higgs recoil to Z Calorimeter: Neutral particles  E/E <½ of LEP Calorimeter: Neutral particles  E/E <½ of LEP good bad W/Z separation W W Z Z Vertex Detector b/c/  tagging Rvtx<1/5 LHC Vertex Detector b/c/  tagging Rvtx<1/5 LHC Improves  (c-tag) 12 July 2007 Akiya Miyamoto 8 DESY Computing Seminar

9. ILD DESY Computing Seminar Challenge of Software Challenge  Enrich ILC physics case, taking progresses of HEP ( work with theorists )  Show the proposed detector can do physics and meet the ILC goal Performance target  Particle Flow :  Very good jet energy resolution by highly segmented calorimeter  Vertexing:  2nd/3 rd vertex reconstrucion  Tracking:  excellent  P/P in ILC condition Studies based on full simulation and realistic reconstruction are necessary  Mokka  Marlin Reconstruction PandoraPFA LCFIVertex Track Reconstruction 9 12 July 2007 Akiya Miyamoto

10. ILD DESY Computing Seminar GLD+LDC  ILD At the time of ILC RDR, 4 detector concepts were considered. For LOI submission in 2009, GLD and LDC agrees to merge and formed the ILD group. In order to define ILD concepts,  studied benchmark processes  checked consistency of software tools  optimized ILD parameters 10 12 July 2007 Akiya Miyamoto LDC: Small cell CAL. Gaseous Tracker 4T European based LDC: Small cell CAL. Gaseous Tracker 4T European based GLD: Small cell CAL. Gaseous Tracker 3T Asian based GLD: Small cell CAL. Gaseous Tracker 3T Asian based

11. ILD ROOT objects : Event Tree & Configuration GLD software tools Beamtest Analysis Event Reconstruction Digitizer Finder Fitter Detector Simulator QuickSim FullSim Event Generator Pythia CAIN StdHep Physics Analysis Jet finder  Link to various tools at http://acfahep.kek.jp/subg/sim/soft  GLD Software at http://ilcphys.kek.jp/soft  All packages are kept in the CVS. Accessible from http://jlccvs.kek.jp/ 12 July 2007 Akiya Miyamoto 11 DESY Computing Seminar

12. ILD JSF Framework: JSF = Root based application  All functions based on C++, compiled or through CINT  Provides common framework for event generations, detector simulations, analysis, and beam test data analysis  Unified framework for interactive and batch job: GUI, event display  Data are stored as root objects; root trees, ntuples, etc development has started since 1999 Release includes other tools  QuickSim,  Physsim(event generators)  BSGen(generate luminosity spectrum)  Analysis utilities  … 12 July 2007 Akiya Miyamoto 12 DESY Computing Seminar

14. ILD Jupiter/Satellites for Full Simulation Studies JUPITER JLC Unified Particle Interaction and Tracking EmulatoR IO Input/Output module set URANUS LEDA Monte-Calro Exact hits To Intermediate Simulated output Unified Reconstruction and ANalysis Utility Set Library Extention for Data Analysis METIS Satellites Geant4 based Simulator JSF/ROOT based Framework MC truth generator Event Reconstruction Tools for simulation Tools For real data Jupiter has modular structure for easy installation of sub-detectors Jupiter can run as a standalone job or a module of JSF Geometry parameters are set by an ascII file read-in at run time Special feature to store pre- and post- point of tracks before/after Calorimeter and break points for PFA studies 12 July 2007 Akiya Miyamoto 14 DESY Computing Seminar

15. ILD ILD Optimization Procedure 15 12 July 2007 Akiya Miyamoto DESY Computing Seminar Whizard Physsim StdHep MOKKA Jupiter LCIO Marlin Sattelites LCIO DST and Analysis LDCGLD StdHep: Same generator data LCIO: Common IO format GLDPrim/LDCPrim: Similar detector model LCIO helps to collaborative works for detector optimization Software inter operativity After ILD optimization, LDC framework was selected as the baseline for LOI studies. No time to really merge GLD and LDC software

16. ILD Optimization by Benchmark Process 16 12 July 2007 Akiya Miyamoto DESY Computing Seminar Using several detector models, performance to separate W/Z in jet mode have been studied using SUSY processes by Taikan Suehara No significant differences are seen

17. ILD Benchmark : 500 GeV  pair 17 12 July 2007 Akiya Miyamoto DESY Computing Seminar Only significant difference among detector models found for  full reconstruction, example in              For reconstruction of both g from  0    Smaller segmentation (5x5mm 2 ) and larger radius advantageous  Impact on physics sensitivity less pronounced Jupiter Mokka Jupiter Mokka

18. ILD ILD Design 18 12 July 2007 Akiya Miyamoto DESY Computing Seminar 3x Dbl. Layer VTX Support of BP/VTX/SIT Forward Component Box support option B=3.5T, R ECAL =1.85 m

19. ILD ILD LOI and beyond 19 12 July 2007 Akiya Miyamoto DESY Computing Seminar LOI was submitted March 2009 Validated by IDAG in September 2009 Next step is to develop Detailed Baseline Design (DBD) by 2012 Re-baseline of ILC, working together with accelerator colleagues

20. ILD GDE Schedule 20 12 July 2007 Akiya Miyamoto DESY Computing Seminar

21. ILD DESY Computing Seminar Software in DBD era Guidelines for software related studies given by Research Director  “Develop a realistic simulation model of the baseline design, including faults and limitation”  “Simulate and study updated benchmark processes including 1 TeV, with background conditions and demonstrate physics capabilities” ILC re-baseline  GDE is updating ILC parameters, taking account R&D progressed since RDR  New parameter will affect beam background conditions and physics performance  software based studies are necessary For ILD  Implementing GLD goodies to LDC and improve LDC soft to meet requirements for DBD  ILDsoft  Improve our tools taking into account lessons in LOI era to meet RD’s request 21 12 July 2007 Akiya Miyamoto

22. ILD DESY Computing Seminar ILD Software tools Generators  Common Stdhep data  Whizard/PhysSim packages Simulation: Mokka  Geant4 application Reconstruction  Marlin Framework  Reconstruction tools as Marlin Processors Core tools  LCIO : standard for persistency format and event data model  Gear, LCCD, CED, …  Grid tools and ilcsoft-install 22 12 July 2007 Akiya Miyamoto  Digitization Simulation: Mokka  Geant4 application

23. ILD DESY Computing Seminar Mokka simulation Core developped by LLR, using Geant4 Geometry data are given by MySQL database  “scalable geometry” has been useful for ILD optimization  Many sub-detector configuration co-exists, even for beam-test detectors. ILD_00 model  ILD_n  fairly detailed geometry Mokka reads  ILC Common Generator samples in stdhep format  GunieaPig beackground particle data  …. Mokka outputs  SimCalorimeter, SimTrackerHits by LCIO 23 12 July 2007 Akiya Miyamoto ILD in Mokka by 3D pdf

24. ILD DESY Computing Seminar VXD in Mokka Two geometries are available in Mokka  Common to DEPFET, FPCCD, CMOS  Cryostat is present, but cables and sensors are not addressed.  to be improved for DBD 24 12 July 2007 Akiya Miyamoto Sensor structure Used for ILD-LOI

25. ILD DESY Computing Seminar Silicon Trackers and TPC 25 12 July 2007 Akiya Miyamoto 4 Silicon trackers in ILD: SIT, FTD, ETD, SET Cylinders/Disks  strip sensors for DBD Geometries in Mokka FTD SIT ETD TPC Gas- Ar/CF4/C4H10 Cu, mylar, G10, Air for Field cage and end plate ( equivalent mass )  No phi-dependence, but OK. TPC is very uniform device

26. ILD DESY Computing Seminar ECAL Mixed readout ( ScEcal/SiECAL ) will be considered in DBD study. 26 12 July 2007 Akiya Miyamoto ScECAL ECAL module side view : incl. dead spaces SiECAL Two readout options, sharing same structure: Silicon and Scintillator 5x5mm 2 Si SiECAL : baseline for LOI, detailed strcture used for LOI study

27. ILD DESY Computing Seminar HCAL Analog HCAL Active: Scintilator 27 12 July 2007 Akiya Miyamoto cylindrical 8/16-sided Cross section of 1 module 1 layer Digital HCAL Active: RPC spacer Float Glass Mylar Graphite PCB Elec. Float glass Graphite Mylar free space RPC ( cross view) Realistic geometry already implemented in Mokka Optimizations : 8/16-sided vs cylindrical & scintillator vs RPC thickness ( # layers ), gaps, tail catchers, absorber materials,

28. ILD DESY Computing Seminar Forward detectors in Mokka Consists of LCAL, BCAL, LHCAL, Beam tube and Masks Mokka model ~ CAD model, but CAD model will evolve with time and Mokka model needs to follow 28 12 July 2007 Akiya Miyamoto ILD CAD model ILD Mokka model LHCAL BCAL New LCAL driver Tile gap FEchips side view of 1 layer Front view

29. ILD DESY Computing Seminar Muon system, Coil and Yoke Updated version of the Muon system has been prepared. Fairly detailed geometry, waiting integration in the central code Magnetic field  For LOI study, uniform Solenoid field for physics study approximated anti-DID field for background study  For DBD Better anti-DID field map is necessary for performance study, at least. Uniform field or realistic anti-DID: –need to consider balance among code readiness, CPU penalty, improvements in reality, …… 29 12 July 2007 Akiya Miyamoto Barrel Muon Tail catcher in cryostat

30. ILD DESY Computing Seminar Cables/Services Cables, services, dead materials for data out/power in/cooling/gas flow sub-detector drivers implements their own materials. To address materials in sub-detector boundaries, small WG has been setup within ILD for  coordination between sub-detectors/optional detectors  defining layout and material budgets  Implementation in Mokka will follow Under new European AIDA framework, new geometry tool kits are in development. New kit will allow consisten geometry treatment among CAD Model, Simulator model, Analysis model. We will be benefitted from this new development 30 12 July 2007 Akiya Miyamoto Mokka: Inner part of ILD ???

31. ILD DESY Computing Seminar Reconstruction Tools Marlin analysis flow  Digitizer for all sub-detectors  Tracking: LDCTracking for TPC and VTX/Silicon trackers  migration to new C++ version in progress  Particle Flow Calorimeter clustering Associate track and calorimeter and create PFObjects ( = primary particles )  Jet clustering  LCFIVertexing : tag each jets  Output as REC data and DST data by LCIO 31 12 July 2007 Akiya Miyamoto

32. ILD Jet Measurements in ILC Det. Typical feature of ILC events and detector e+e-  Z 0  q qbar  E/E(  p/p) p(GeV/c)/E(GeV) HD CAL  E/E=50%/√E Tracker （ TK)  p t /p t =5x10 -5 p t EM CAL  E/E=15%/√E Resolution of a ILC detector Principle of PFA detector Charged particles by tracking device Remove charged particle signals in CAL ( avoid double counting ) Large bore magnet, Large B-field Highly segmented CAL to separate clusters by charged and neutral particles.  Patten reconstruction is a key. Large bore magnet, Large B-field Highly segmented CAL to separate clusters by charged and neutral particles.  Patten reconstruction is a key. 12 July 2007 Akiya Miyamoto 32 DESY Computing Seminar

33. ILD DESY Computing Seminar Pandora PFA 33 12 July 2007 Akiya Miyamoto Originally developed by Mark Thomson (U. Cam) as a Marlin Processor. V3.2 was used for ILD LOI. Used by SiD ( thanks to common LCIO format ), achieved  E/E ~ 25%/√E for Z pole jets Now re-organized to PandoraPFANew, as a stand alone package PFA algorithm slide by John Marshal (ILD Soft WS)

34. ILD DESY Computing Seminar LCFIVertexing LCFI group developed LCFIVertexing package. Apply algorithm for each jets It consists of two parts,  ZVTOP/ZVKIN : Find vertexies from probability of overlapped trajectories  NeuralNet for tagging and vertexing Jet with 1 vertex (=IP) –may contain 1 displaced track (D) > 1 vertecies: –Pt corrected vertex mass is a very good variable to identify quark floavor. –Other variables (joint-track probability, etc.) LCFIVertexing also output  vertex charge: powerful discriminator of b and anti-b quarks Used by both SiD and ILD ( thanks to LCIO ) 34 12 July 2007 Akiya Miyamoto Pt corrected vertex mass

35. ILD DESY Computing Seminar LCFIVertexing Typical b/c tag performance With neural nets tuned for Z  qq events Very good performance ( also thanks to the vertex detector placed very close to IP) Issues to be studied  performance with beam backgrounds  performance in multi-jet environments 35 12 July 2007 Akiya Miyamoto Vertex charge of b-jets in ttbar events Eff. ~ 28% with purity 75% for a b-jet, incl. b  B 0

36. LCFIVertex Apply neural net analysis for each jets. Jets with no displaced vertex, but >1 misplaced track.  C-jets ( charged D meson)  2 misplaced track : b  c jets  joint probability of all track to originate from primary vertex  More than one vertex found  decay length and decay length significance  momentum of tracks assigned to vertecies  pt corrected vertex mass  number of tracks of non-primary vertex  secondary vertex probability  joint probability of all tracks to originate from primary vertex

37. GRID GRID is an infrastructure for a large scale international researches GRID provides  Resources for Large scale computing Large scale data storage  International/Inter-regional communication basis GRID have been used extensively in ILD LOI studies  for MC productions  Data sharing between Japan – Germany/France/UK

38. Tohoku Univ. KEK Univ. of Tsukuba Nagoya Univ. Kobe Univ. Hiroshima IT Network in Japan and GRID Major HEP projects: – Belle, J-PARC, ATLASongoing projects – ILC, Belle2future projects Also covering – Material science, bio-chemistry and so on using synchrotron light and neutron source – Radiotherapy as technology transfer KEK has a role to support university groups in these fields. – including Grid deployment/operation. Hiroshima Univ. (Alice) U of Tokyo (Atlas)

39. SINET3 39 12 July 2007 Akiya Miyamoto DESY Computing Seminar http://www.sinet.ad.jp/ Round Trip Time: KEK  IHEP/KISTI ~ 100msec  FNAL ~ 200msec  DESY/IN2P3 ~ 300msec

40. GRID infrastructures 40 12 July 2007 Akiya Miyamoto DESY Computing Seminar Middle waregLiteNAREGIGfarmSRBiRODS Belle (Belle2)UsingPlanningUsing AtlasUsing Medical AppsUsingDevelopingPlanning ILCUsingPlanning J-PARCPlanning Testing LCGRENKEI KEKCC supports both LCG and NAREGI/RENKEI Many Japanese HEP groups are joining LCG NAREGI middleware is being deployed as the general purpose e-science infrastructure in Japan RENKEI is developing a system to provide a seamless user environment between the local resources and multiple grid environment

41. GRID for ILC Two Vos have been used:  CALICE-VO: Test beam data analysis and MC. Standard data processing in GRID  ILC-VO: Needs huge CPU resources for the studies. Available only on GRID Standard MC samples ( ~ 50TB) are on GRID for sharing Status:  A typical data transfer rate from IN2P3/DESY to KEK: ~ 200kB/sec/port a frequent timeout for transfer of ~ 2GB: Cured by removing a time out at IN2P3  Overhead of catalog access ILD DST: many small size DSTs, limited by CPU time for a MC job. MC and DST production at DESY/IN2P3  Merge DSTs to create a large size file, then replicated to KEK

42. A typical GRID performance File transfer: IN2P3  Kobe, 184 files/210 GB in 13 hours - part of ILD LOI study, in Dec. 2008 - 10 ports/job Pedestal in transfer time ~ 20~60sec.  < 100MB is not effective. Instantaneous transfer rate: average 4 MB/sec, Max. 10 MB/sec  not great, but has been used for real works successfully Data size vs TimeTransfer rate During Dec. ‘08 to Feb. ’09, O(10TB) data have been exchanged through GRID. It was crucial for the successful LOI studies. During Dec. ‘08 to Feb. ’09, O(10TB) data have been exchanged through GRID. It was crucial for the successful LOI studies.

43. ILD DESY Computing Seminar Recent issues on GRID After LOI, KEK has extended GRID resources  CPU: 0.3M SI2K to 6M SI2K ( utilize old CPU resources )  Storage: IBM HPSS as the backend storage. Tape capacity up to 3 PB. shared by batch server and many other groups. Operational issues  network speed outside Japan  With increasing WNs, many new problems seems to appear Files in HPSS : file transfer breaks frequently Access to MySQL server from WN Disk space of WNs are not sufficient for storing large temporary data Over-loaded WMS Very slow turn around from job submission to job out retrieving  Failure rate is not low enough and system tuning is yet to be done. 43 12 July 2007 Akiya Miyamoto

44. ILD DESY Computing Seminar Summary Very extensive developments and studies have been done using ILC software tools. We are aiming to produce DBD by 2012 and software based study will play crucial role in it. Our main efforts right now are  updates of simulator models  improvements of core software tools Improvements of reconstruction tools and new benchmark studies will follow soon 44 12 July 2007 Akiya Miyamoto