GLAST LAT ProjectSAS PDR, Aug. 17, 2001 T. Burnett1 Infrastructure Sources Simulation Reconstruction

GLAST LAT Project SAS PDR, Aug. 17, 2001

T. Burnett 1

InfrastructureInfrastructureSourcesSources

SimulationSimulationReconstructionReconstruction


T. Burnett 2

The Processing chainThe Processing chain

SourceFluxes

Geometry

ParticleTransport

“Raw”Data

Recon

BackgroundRejection

-Particle ID

SourceFluxes

Geometry

ParticleTransport

“Raw”Data

Recon

BackgroundRejection

-Particle ID

Real Data


T. Burnett 3

Our Products: much more than code!Our Products: much more than code!

• Support infrastructure, must support a variety of clients:– developers– sophisticated users– end users

• Elements:– Supported platforms & compilers– Development environments– Coding and documentation standards– Build tools– Framework– Analysis tools


T. Burnett 4

Basic principles for technology choicesBasic principles for technology choices

• Don’t invent anything unnecessarily• Borrow from existing solutions, experience

High energy physics – very similar parameters: detectors, analysis requirements,

data, users– Pioneer was here at SLAC: the Babar experiment in mid 90’s,

• Broke with Fortran-oriented past: unix, OO C++

• Adopted industry-standard CVS for version management

• Invented package-oriented build system SRT

• Invented a framework for managing processing steps

• Successfully trained physicists to deal with new environment


T. Burnett 5

Technology choices: languageTechnology choices: language

• Object-oriented C++– Basic value of encapsulation of data now well-established– Build on success of Babar and all other new HEP

experiments: Belle, D0, CDF, ATLAS, CMS, LHCb– Now a standard, most compilers approach this– Standard Template Library provides rich menu of algorithms

and containers.– Required to use a framework


T. Burnett 6

Technology choices: platformsTechnology choices: platforms

• Windows PC– Not a choice for most of the HEP experiments

– Our preferred development environment due to rapid development made possible by Microsoft Visual C++ MSDEV

• linux– The preferred choice for European developers

– Needed for SLAC batch support

• solaris– May be needed for SLAC batch.


T. Burnett 7

Technology choices: code versioningTechnology choices: code versioning

• CVS!– Concurrent Versions System, the dominant open-source

network-transparent version control system. – Useful for everyone from individual developers to large,

distributed teams: • Its client-server access method lets developers access the

latest code from anywhere there's an Internet connection.

• Its unreserved check-out model to version control avoids artificial conflicts common with the exclusive check-out model.

• Its client tools are available on most platforms.


T. Burnett 8

Choices: Code managementChoices: Code management

Legacy of Babar’s SRT: building apps from packages– Collection of source files, with public header files in a folder

(usually) with the package name– Produces a binary library and/or executable

CMT (for Code Management Tool): our choice– Developed in response to deficiencies of SRT, adopted by

LHCb and ATLAS– Supports Windows– Clean model for package dependencies

• Support for compile-time, link-time, and execution-time

– Configuration specified in a single file– Includes tool to generate makefiles, or MSDEV files


T. Burnett 9

Choices: frameworkChoices: framework

• Requirements:– Support event-oriented processing:

• initialization

• loop over generating or processing events

• termination

– Flexible way to specify processing modules to be called in loop

– Provide services, especially for making n-tuples and histograms


T. Burnett 10

GaudiGaudi

Open source

Stable, but active developers

Very good documentation

All code called via component interfaces:– Algorithm

– Service

– Converter

– DataObject

Job control parameters set in job options file.

The Gaudi Framework for LHCb, showing package dependencies (GLAST is similar)


T. Burnett 11

Gaudi algorithmGaudi algorithm

• Components are similar to Corba or COM: implement an abstract interface.

• Easy to substitute components• Example diagram: A

ConcreteAlgoritm:• Implements 2 interfaces• requests services from 6

services via abstract interfaces


T. Burnett 12

Data flow in the Gaudi frameworkData flow in the Gaudi framework


T. Burnett 13

Choices: I/O and analysisChoices: I/O and analysis

ROOT


T. Burnett 14

DocumentationDocumentation

• Gaudi, CMT, cvs: user guides available• Local guides (web-based)

Software

[Getting Started with GLAST Software (Your How-To Page) ] [Web Access to CVS repository] [Using GlastSim] [Using tbsim] [Using ROOTWriter] [Using tb_recon]

Support [Whom to Call??] [Facilities at SLAC] [UW Windows Server]

Projects[GAUDI] [GEANT4] [PDR] [Software PDR] [Event Display]

Other Software Resources

[Italy] [UCSC TB Recon] [Goddard] [NRL Software] [Hiroshima]

Tools

[Telecon VRVS] [Using VRVS for Glast][Instant Msg ICQ] [Using ICQ for Glast][CVS] [Using Cvs for Glast][CMT] [Using CMT for Glast][Root] [Using Root for Glast] [Root at FNAL][Creating PEGS files]


T. Burnett 15

Help in the form of a GUIHelp in the form of a GUI


T. Burnett 16

Visual CMT (VCMT)Visual CMT (VCMT)

• GUI interface to:– CMT: manage packages – CVS: check out, commit – MSDEV: build, or start its GUI


T. Burnett 17

Coding documentation, standardsCoding documentation, standards

• Inline documentation• Standards


T. Burnett 18

Managed setups for developers Managed setups for developers

University of Washington Terminal Server– Uses Windows 2000 Terminal

Server: free clients available for any Windows operating system

– Complete environment available for users, including VCMT

– etc.

SLAC unix– Standard group .cshrc

– etc.


T. Burnett 19

Sources: Incident FluxSources: Incident Flux

• Service to provide incoming particles for simulation

• Types that must be available:– Primary and secondary

Galactic Cosmic Rays: protons and electrons

– Albedo gammas– gammas for testing

resolution– Galactic gamma sources

• distributions of energy spectra

• angles with respect to:– local zenith– spacecraft – galaxy

Flux Service: • Selects from library (XML spec)• Manages orbital parameters• Returns particles generated by

selected sourceSelected Source: return particles

depending on orbit


T. Burnett 20

Rootplot: A useful utility to study sourcesRootplot: A useful utility to study sources

Plot at right generated by a utilty program in the flux package.

Can choose any combination of sources described in the XML file, and generate distributions of energies and angles that would be provided to the service.

Plot of the energy spectra for various components of a proposed background mixture,

including: :•chimeavg, representing a average rate for the CHIME model of primary proton cosmic rays;• albedo_proton, the spectrum of albedo and reentrant protons corresponding to recent measurements;• albedo_gamma, secondary gammas from the horizon, and• CrElectron, a mixture of primary and secondary electrons and positrons. The abscissa is the kinetic energy of the particles (gamma, proton, or electron) in GeV, and the ordinate the flux times energy integrated over angles, in particles/(m2 s).


T. Burnett 21

Detector GeometryDetector Geometry

Interactive 3-3 display is vital

GUI also can control events


T. Burnett 22

Depositing energy: bookkeeping designDepositing energy: bookkeeping design

• Particles transported by the simulation deposit energy in matter by ionization loss, in many small steps

• Each loss is associated with the given volume, two strategies– Single-step: every step saved– Volume integrating: only keep

total, perhaps in subdivisions

• Primary objective: create realistic detector response

• Secondary objective: preserve enough information about the underlying event to guide design and evaluation of reconstruction strategies– Parent particle: incoming or e+/e-

from pair conversion

3 GeV photon interaction (charged particles shown only)

Detector responses shown


T. Burnett 23

Digitization RequirementsDigitization Requirements

• ACD– total energy deposited

– position of all steps and associated MC parent particle

• TKR– the dead material energy loss

must be segmented at least by plane

– Silicon treated as one volume, but complete detail of each step in the silicon.


T. Burnett 24

Digitization RequirementsDigitization Requirements

• CAL– A calorimeter crystal, or “log”

will be treated as a single volume for the simulation, but energy deposition will be segmented to allow the light collection digitization stage to deal with the distribution of energy throughout the log. In addition, it is planned to register energy sum and energy-weighted longitudinal position moments.

– Depending on the readout mode, the best or all four PIN diode readouts are simulated and include the light output taper from end to end of the logs. Future upgrades include electronic non-linearities and optical gains


T. Burnett 25

Event ReconstructionEvent Reconstruction

• takes the raw readouts from the detector elements, converts them to physics units (e.g. energies in MeV, distances in mm)

• performs pattern recognition and fitting to find tracks and then photons in the tracker

• finds energy clusters in the calorimeter and characterizes their energies and directions

• uses the ACD to allow rejection of events in which a tile fired in the vicinity of a track extrapolation


T. Burnett 26

Tracker ReconstructionTracker Reconstruction

• Pattern recognition in x and y planes

• 2-d Fitting• Associate x, y tracks• 3-d fitting with Kalman filter


T. Burnett 27

PSF estimation from track fitPSF estimation from track fit

Angle between reconstructed and incident photon (radians)


T. Burnett 28

Calorimeter ReconstructionCalorimeter Reconstruction

• The calorimeter consists of 16 modules of 8 layers of 12 CsI(Tl) crystals in an hodoscopic arrangement, this is to say alternatively oriented in X and Y directions, to provide an image of the shower. It is designed to measure energies from 30 MeV to 300 GeV and even 1 TeV.

• However, the calorimeter is only 8.5 X0 thick and therefore cannot provide good shower containment for high energy events, though these events are very precious for several astrophysics topics. Indeed, the mean fraction of the shower contained can be as low as 30% at 300 GeV, normal incidence. In this case, the energy observed becomes very different from the incident energy, the shower development fluctuations become larger and the resolution decreases quickly.

• Two solutions have been pursued so far to correct for the shower leakage. The first solution to correct for the energy loss is to fit a mean shower profile to the observed longitudinal profile. There are 2 free parameters, E0 and the starting point of the shower to take into account early fluctuations.

• The profile fitting method proves to be an efficient way to correct for shower leakage, specially at low incidence angles when the shower maximum is not contained. The resolution is 18% for on axis 1 TeV photons, which is a 50 % improvement compared to the raw sum of the energies recorded in the crystals.

• The second method uses the correlation between the escaping energy and the energy deposited in the last layer of the calorimeter. The last layer carries the most important information concerning the leaking energy: the total number of particles escaping through the back should be nearly proportional to the energy deposited in the last layer. The measured signal in that layer can therefore be modified to account for the leaking energy.

• The methods presented significantly improve the resolution. Up to 1 TeV, the resolution on axis is better than 20 %, and for large incident angles (more than 60 degrees) it is around and even less than 4 %. It should be noted that the best layer correction is more robust since it doesn’t rely on a fit, but its validity is limited to relatively well contained showers, making it difficult to use at more than 70 GeV for low incidence events. There is still some room for improvements, especially by correcting for losses between the different calorimeter modules and through the sides.

Documents

GLAST LAT ProjectSAS PDR, Aug. 17, 2001 T. Burnett1 Infrastructure Sources Simulation Reconstruction