1 Status of Software Tools Akiya Miyamoto KEK 5-March-2005 ILC Detector Workshop Based on acfa-sim-j activity.

1

Status of Software Tools

Akiya MiyamotoKEK

5-March-2005

ILC Detector Workshop

Based on acfa-sim-j activity

2A.Miyamoto , ILC Detector Workshop (3-5, March, 2005)

Contents

Status of Simulation and Analysis tools

A brief review of tools in Europe and N.A.

Summary


Objectives of Softwares

Physics studies Event generators Fast detector monte carlo

Detector studies Geant3 Geant4 Reconstruction: clustering, track fitter, vertexing, …

Beam test studies Data storage and analysis Simulation

Communication / Information


Goals of in FY05

By LCWS2004, we had JSF framework and Quick Simulator for physics studies Jupiter with basically CDC, IT, VTX, and IR. basic structures of Satellites/Uranus

Goals of studies in FY05 are detector optimization based on Full detector simulation. Implement “GLD” geometry in Jupiter Study PFA performances by an ultimate condition

Implement “tower” calorimeter Develop analysis tools

Study physics performance vs detector choice.


Organization of tools

All packages are kept in the CVS repository at jlclogin.kek.jp.

Package organization lcbase leda : Analysis tools (Kalman fitter, Anlib) jsf : Framework, StdHep, LCIO lclib : “GLD” parameter for QuickSim physsim : Helas-based generator – phythia6 compat Jupiter : Full simulation based on Geant4 Uranus : Data analysis packages Satellites : Data analysis packages for MC data

Recently, we reorganized packages to reduce dependances among packages.


Jupiter/Satellites Concepts

JUPITERJLC Unified

Particle Interactionand

Tracking EmulatoR

IO Input/Outputmodule set

URANUSLEDA

Monte-Calro Exact hits ToIntermediate Simulated output

Unified Reconstructionand

ANalysis Utility Set

Library Extention for

Data AnalysisMETIS

Satellites

Geant4 basedSimulator

JSF/ROOT basedFramework

JSF: the analysis flow controller based on ROOT The release includes event generators, Quick Simulator, and simple event display

MC truth generator Event Reconstruction

Tools for simulation Tools For real data


Progress of Jupiter since 8th ACFA

Initialization of Calorimeter geometry: ~3min for 4cmx4cm tile 　 ~20 sec. even for 1cmx1cm tile. 　 ~consume about 500MB memory.

Problem of HD Cal response for shorter range cut. Caused by an artificial cut on number of steps and fixed

Energy deposit in EM Cal depends on the range cut. Larger effect for thinner scintilator

MUDRefined geometry and implemented sensitive detector

Update of IO interface Write/read >2 Gbyte files OK StdHep interface is provided through JSF LCIO output class is prepared, needs test

MCParticle and exact hits of VTX, IT, TPC, CAL, MUD are written.

Y.Yamaguchi

S.Yamamoto

K.Fujii, S.YamamotoJeri, T.Takeshita

H.Ono

A.Miyamoto


Geometry in Jupiter

Solenoid

Hadron Calorimeter

Muon/Iron

Elemag. Calorimeter IT VTX

Forward Cal.

QC1

TPC


Standard Geometry of Jupiter

Solenoid

TPC

Hadron Calorimeter

Muon/Iron

IT VTX

Forward Cal.

QC1

Elemag. Calorimeter

As of Nov. 4By H.Ono(Niigata)

GLD


Status of Metis

Current aim is to prepare a minimum set of Metis modules for studies of Particle Flow Algorithm.

Novice users will be able to do physics analysis using information of PFO classes.

As a first step, a cheated track finder and a cluster maker, etc are in preparation in order to know ultimate performance.

Each module is independent, thus shall be easy to implement different reconstruction algorithm according to interests


make smeared TPC hits from exact hit

make tracks from TPC

make hybrid tracks ( TPC+IT+VTX)

make smeared/merged CAL hits from exact hit

make cluster from CAL hits

make Particle Flow Objects

jet clustering

Jupiter result

Physics study

Metis Analysis Flow


Cheated PFO analysis

ZH event at Ecm=500 GeV

By K.Fujii(KEK), S.Yamamoto(GUAS), A.Yamaguchi(Tsukuba)

- Exact hit points of TPC and CAL are displayed.-Hits belong to the same PFO are shown with the same color

-A framework of event display in JSF is used.

X3DROOT’s 　 X3d view of the same event

X3D-JetSame event, after a forced 4-jet clustering on PFObjects


Satellites Hit maker for IT, VTX included Hybrid Track fitter for TPC/IT/VTX are prepared

Fixed many memory leaks and segmentation violations. Now jet energy studies using higher statistics sample is possible.

Progress of Satellites since 8th ACFA WS

A.Yamaguchi, T.FujikawaK.Fujii, Nagamine


Jet mass resolutionBy S.Yamamotoat G4 workshop

M/M~36%/Sqrt(M)

Results by T.TakeshitaPythia+Smearing

(Mw) is about same as Jupiter result

Jupiter result has long tail. Why ? Insufficient coverage ? Bad performance in the forward region ? Other factors, such as V0, Kink ? Event selection ?

The goal, 30%/Sqrt(E), is not satisfied. Better segmentation for EM/HD ? Better energy resolution of EM/HD ? 50%/Sqrt(E) is not as good as JLC/GLC detector.

How about other processes/ECM

Brief review of simulation tools in other regions

Brief review of simulation tools in other regions


Software tools in EU

Generator

geometry

AnalysisRecon-structionSimulation

Java, C++, FortranGeant3, Geant4

Java, C++, FortranJava, C++, Fortran

LCIO Persistency Framework

Brahms (geant3)Brahms (geant3)

SIMDETJAS3 & Wired4

Mokka (Geant4) Marlin (started)


Mokka Overview geant4 based full detector

simulation for the ILCD

developed at LLR-Ecole Polytechnique (P. Mora de Freitas, G. Musat)

http://polywww.in2p3.fr/geant4/tesla/www/mokka/mokka.html

some features: steering files for configuration all geant4 physics lists available writes LCIO reads StdHep / ASCII HepEvt

Geometry

MySQL databases geometry parameters one database per subdetector

C++ Geometry Drivers one for each subdetector type

(e.g. TPC, HCAL) define material and sensitive d

etector

abstract geometry layer: CGA

Data base can define ILCD geometry as well as test beam setups.













LCIO

LCIO SW-Architecture

*.slcio files (SIO)

common API

LCIOC++ implementation

LCIOJava implementation

Java API C++ API f77 API

JAS/AIDA root hbook

compressed records,pointer retrieval

generated fromone sourceusing AID


LCIO Data Model


Software tools in US

Evolution of simulation software

Gismo

Hep.lcdReconstruction

SLIC

org.lcsim

LCDRoot

LCDG4

LCS

Simulation Architecture Proposal

LCDD SLIC

LCIOGDML

LCIO Events

Geant4

Stdhep Events

LCDD XML

LCCDD XML

GeometryConvertor

org.lcsimReco

ILC Simulation Architecture Proposal

reads

createsreads

usesuses usesuses

reads reads reads+writes

creates

uses

data application

uses

action

J.I. McCormick, SLACGDML – Geometry Description Markup Language

LCIO – Linear Collider I/O

LCIO Events – LCIO event collection

LCDD – Linear Collider Detector Description

SLIC – Simulator for the LInear Collider

LCDD XML – XML file conforming to lcdd.xsd schema

Geometry Convertor – converts compact description to LCDD

LCCDD – Linear Collider Compact Detector Description

Stdhep Events – binary encoded HEPEVT

Geant4 – Simulator framework

Java Reco – Java-based reconstruction framework


Simulation tools summary

N.A. EU Asia

Generator format StdHep/HepEvt StdHep/HepEvt StdHep/HepEvt/CAIN, ….

Geometry Description

G4 GDML+ MySQL+XML Hard coded XML

Fast Sim FastSim SimDet QuickSim

Base System JAS3+AID JAS3+AID+Root Root

Framework LCDROOT/JAS ? MARLIN JSF

IO LCIO LCIO ROOT+LCIO

Full Simulation LCDG4SLIC Mokka Jupiter

Beam test Sim LCDG4SLIC Mokka independent

Reconstruction Org.lcsim(Java) BrahmsMarlin Uranus/Satellites

Language JAVA/C++ JAVA/C++/FORTRAN C++


Reconstruction tools

Vertexing ZVTOP V0 finder, Kink Finder

Track finder/fitter Khalman fitter

Particle Flow Analysis Track based or Calorimeter based or ….

/ e / / / jet finder Infrastructure

Geometry data base Visualization

International cooperation is important to reduce un-necessary overlaps of efforts. - LCIO is the first step- The second step, what and how ?


Summary : Next step

Uranus/Satellites for perfect PFA is ready. Let’s study, Jet mass/energy resolution vs detector configuration, such as

CAL cell size and material Develop real reconstruction tools, for

PFO analysis Track finder for TPC Track finder/fitter for TPC Packages for particle ID, such as , e, , , jet Vertex finder/fitter

Improvement of Jupiter Put more realistic models of detector, especially massive parts

of inner components. By when ? At what precission ? Can we use XML tools ?

Urgent task: Realistic models of CAL, Forward Tracker, and IR. Magnetic field for Return Yoke, QC, ..


Next step - 2

Infrastructure Physics and background generator samples

What processes ? Who will prepare ? How to exchange ? Creation of simulation data

100K ZH events at 300GeV, – CPU time ~100 days (1 Xenon 3G)– Data size : lcio ~60GB (sim. exact hits only)

jsfj4 ~300GB (includes additional info.) How to produce ? Who will produce ? How to exchange ? Do we need GRID ???

“Database” for detector geometry Jupiter Sattelites Analysis ; should avoid hand-written data A common among detector concepts would be preferable

– MySQL : DESY and KEK, SLAC: Oracle– Geometry model/interface is not clear– How to realize open access world wide ?

1 Status of Software Tools Akiya Miyamoto KEK 5-March-2005 ILC Detector Workshop Based on acfa-sim-j activity.

Documents