HCAL DQM Software generically can be divided into 2 components: CMSSW and non-CMSSW. CMSSW incorporates all the Analysis Tools, handles the Monitor Elements Collections(booking, resetting, however NO DRAWING!), provides generic functionality to monitor the status of HCAL and its every subsystem. Non-CMSSW Component can be further divided into plugins for the DQM GUI and generic HCAL DQM scripts that provide some functionality for the Local Detector Diagnostics(analysis of the Local Runs), handling of some data copying, etc... Therefore, for the purpose of the description, we differentiate between CMSSW and non-CMSSW components of the HCALDQM.

See HowTo Section for the usage.

HCAL DQM CMSSW Modules/Packages can be divided into:

• Plugins/Sources(a la DQMEDAnalyzer)
• Clients(a la DQMEDHarverster)
• HCAL DQM Application Config Files - which are typically being run with cmsRun

Reside in either DQM/HcalTasks or DQM/HcalMonitorTasks. The first one contains the most recent HCAL DQM Tasks that utilize new functionality in DQM/HcalCommon. The later one contains old HCAL DQM Monitor Tasks that have been during Run I, LS1, and will partially be used during Run 2. These two are written by 2 different groups of people, however provide similar functionality. Here the preference is given to the newest modules residing in DQM/HcalTasks. For further reference regarding the old modules, consult here https://twiki.cern.ch/twiki/bin/viewauth/CMS/HcalDQM.

A list of available tasks(in DQM/HcalTasks):

• Raw
• Digi
• RecHit
• TP
• Timing
• LED
• Laser
• Pedestal

Reside in either HcalMonitorClient or HcalClients. For the timebeing, only HcalMonitorClient is being used for Determining the Status of HCAL. I want to make it clear from onset, there is only 1 real DQMEDHarvester and several simple classes that are being called from it. This is done because these clients are run in sequential order always and also becasue that is the easiest way to generate a summary. That is for the Old Clients. Typically these clients aren't used for any offline analysis if someone wishes to use HCAL DQM Modules, cause they need the histograms themselves, not the status.

A list of clients used to generate HCAL Summary/Status:

• DeadCellMonitor
• HotCellMonitor
• RecHitMonitor
• DigiMonitor
• RawDataMonitor
• TrigPrimMonitor
• NZSMonitor
• BeamMonitor
• CoarsePedestalMonitor
• DetDiagTimingMonitor
• Summary

Not all of them are actually used, but listed nevertheless...

Reside just in the root folder of the repo. Here they are:

• new_hcal_dqm_Global.py
• new_hcal_dqm_Calib.py
• new_hcal_dqm_Local.py

Names should obviously reflect the usecase of these config files. Global is what is being run online, calib is what is being run during the Abort gap as DQM uses DQMCalibrationStream for that purpose. And finally for the local detector diagnostics we run Local. Each file is well documented and should be easy to follow. At the beginning of each there are some bolean variables that help transition from lxplus to P5 or select a map from DB...

To be Provided

As of 09/08/2015, the PRs to include these new modules and new configs have been submitted, however not merged. Therefore some tweaking will be neccesary to run everything. The assumption is that you are goin to run on lxplus

Current Working Branches for Use:

• master-74X for CMSSW_7_4_X
• master-75X for CMSSW_7_5_X

1. cmsrel CMSSW greater or equal than 74X and setup the environment
2. git cms-addpkg DQM/Integration
3. You can either just clone the repo or actually fork it...
4. You will have to copy the neccessary folders from the cloned/forked repo to the location of your cmssw release, into the DQM/ folder. Just like DQM/Integration.
5. Copy the hcal_dqm_* files into the DQM/Integration/python/test
6. Build CMSSW: scram b -j 8 from the src dir of the CMSSW
7. To Run new_hcal_dqm_Global.py on lxplus, you have to set OfflineGT to True, useFileInput to True and put the right tag in.
8. In DQM/Integration/python/test there is a file fileinputsource_cfi.py, which contains the definition for the Pool Source. Put in there the filenames you want to run over. This step is exactly the same as for any analysis - you have to provide the input files...