L1Browser Analyzer Walkthrough

Introduction

The purpose of this tutorial is to provide an walkthrough for the L1Browser analyzer, and also to provide some basic information on how to select the data set read in for this analysis. This analyzer reads in data and does the following:

Set Up Your Environment

First, set up your CMSSW_2_0_5 directory and your path to the CVS repository by typing the following commands into your home directory:

export SCRAM_ARCH=slc4_ia32_gcc345
. /raid/raid4/pg/app/cmssoft/cms/cmsset_default.sh
export CVSROOT=:pserver:anonymous@cmscvs.cern.ch:/cvs_server/repositories/CMSSW
scramv1 project CMSSW CMSSW_2_0_5
cd CMSSW_2_0_5/src

Now, you will need to set up your runtime environment. This command is different if you are working in C-shell or bash. Type the correct command from below while in your src directory:

eval `scramv1 runtime -sh` - if you are using bash
eval `scramv1 runtime -csh` - if you are using C-shell

Log into the CVS repository by typing the following command:

cvs login (use the password "98passwd")

Several packages must be checked out for the CVS repository for the analysis package to run. Check these packages out using:

cvs co -r CMSSW_2_0_5 EventFilter/CSCTFRawToDigi
cvs co -r CMSSW_2_0_5 L1Trigger/CSCTrackFinder

Currently, the L1Browser analysis package is not committed to CVS, so it must be downloaded separately. The analysis package can be found here: http://tier2.ihepa.ufl.edu/~bjack3/Backup/L1Browser_5_26_08_v2.tgz

When that file has been downloaded, copy it to your working directory and unpack the gzipped tar file by executing the following commands:

cp L1Browser_5_26_08_v2.tgz CMSSW_2_0_5/src
tar xzvf L1Browser_5_26_08_v2.tgz

You now have all the files necessary to run the L1Browser analyzer, so now you should compile and build the code using:

scramv1 b

You are prepared to begin analyzing data!

Analyze Data

In this section of the tutorial, you will perform an analysis job on data from the May 2008 CRUZET Global runs. You will need to create a digi file. This is done by executing the following commands:

cd EventFilter/CSCTFRawToDigi/test
cmsRun Write_Digis.cfg

If this ran successfully, you should now have a file called "digis.root" in your test directory. This file contains track finder digis that will be read by the L1Browser analyzer.

Now, run the analyzer on this data using this command:

cmsRun L1Browser.cfg

This job will create a file called analysis.root in your test directory. This file contains all the histograms generated by the L1Browser analyzer. While it is possible to view these histograms through a TBrowser within root, at the time this tutorial was written, the best way to view these plots is to use a script to print each the plots to files. This script prints the generated histograms to *.png files. Run the script by executing the following command.

./PrintPlots.sh

After running this script, a new directory, called "Plots" has been created in your test directory, and contains the histograms printed by the script. If possible, the suggested method of viewing these plots is through a web browser. A web page has been created that allows one to brows these plots in a more organized fashion. When working on the UF analysis cluster, this website can be generated by running the following script:

toWeb.sh TITLE

where TITLE is the name of the run in question. If you are not working on the analysis cluster, you must modify the toWeb.sh script to point to the path that is viewable from the web

If a web area is not available, it is still possible to view these images with an image viewer, such as gv.

You are now able to view the histograms produced by the L1Browser analyzer!