Combining all groups of HLT triggers in a Global Table

Complete:

The HLT runs off raw data files - specifically, off the single FEDRawDataCollection data structure; it can NOT be run off RECO or AOD files which do not keep this data structure. The HLT algorithms perform (regional) reconstruction as needed, and make trigger decisions. The list of HLT path names is reported on the TriggerTables page.

How to run the production HLT within CMSSW

This part of the wiki describes the instructions on how to run the HLT contained in the CMSSW release you are using; no access to ConfDB is required. This mode is typically used for large-scale Monte-Carlo event production.

Preparing a working area

The generic procedure to run the HLT in your favorite CMSSW release, for example CMSSW_5_3_X (run-1 legacy release), CMSSW_7_4_X (release for HLT for the 2015 pp data taking period), CMSSW_7_5_X (release for HLT for the 2015 PbPb data taking period), CMSSW_8_0_0 (release for HLT for the 2016 data taking period), CMSSW_9_2_0 (release for HLT for the 2017 data taking period), CMSSW_10_1_0 (release for HLT for the 2018 data taking period), or any later CMSSW release, is as follows (written here for the 2018 pp data taking release):

cmsrel CMSSW_10_1_11_patch1
cd CMSSW_10_1_11_patch1/src
cmsenv
git cms-addpkg HLTrigger/Configuration
scram build
cd HLTrigger/Configuration/test
./cmsDriver.csh              # if you want to execute the shell script which creates various ready-to-run cfg files

NOTE The above set of commands is sufficient if you only want to run the HLT menus which come with the release you are using. If you need more, or more recent, for example for trigger development beyond 2018, then you need to use the more extended set of instruction listed below.

Using frozen Run-1 or Run-2 trigger menus

CMSSW releases typically contain several HLT menus inside HLTrigger/Configuration, at a minimum a menu for proton-proton collisions (called GRun), a menu for lead-lead heavy-ion collision (called HIon), and a menu for proton-lead collisions (called PIon). Depending on release, these menus are either for the run-1 or the run2 data taking period. For run-1 studies, you should use the run-1 legacy release CMSSW_5_3_23. For run-2 studies you need to used: for 2015 the most recent CMSSW 74X release, for 2016, the most recent CMSSW 80X release, for 2017, the most recent CMSSW 90X.

A few checkpoint pp HLT menus used during LHC Run-1 are frozen and put into recent 53X, 62X, 70X, 71X, 72X, 73X CMSSW releases (in addition to the GRun, HIon and PIon menus discussed above). In the Run-1 legacy release CMSSW 53X, you'll find the 2011 5E33 menu and the 2012 menues 5E33v4, 7E33v2, 7E33v3, 7E33v4 and 8E33v2. In the later CMSSW releases, 62X, 70X, 71X, 72X, used for the preparation of run-2, we only keep the most recent 2012 pp menu, 8E33v2 (aka 2013) or 2014 (including half-rate paths), as a frozen menu. From CMSSW 73X onward, we have dropped all Run-1 menus. The 2015 menus are available in 74X (incl. the ppRef5TeV menu) and 75X (including the ppRef5TeV menu and the PbPb menu). In 2016 we use 80X. In 2017 we will use 90X.

Ready-to-run cmsRun cfg files such as RelVal_HLT_GRun.py are produced using cmsDriver: simply execute the shell script ../test/cmsDriver.csh to create various workflow cfg files. (The cmsDriver parameter syntax to select a specific HLT menu is -s HLT:GRun; using -s HLT only defaults to the GRun menu. Also these frozen menus are accessible via the cmsDriver command-line, see the script cmsDriver.sh in HLTrigger/Configuration/test for examples.)

For example, edit the pp config file RelVal_HLT_GRun.py to use your favorite raw-data input file and run it with cmsRun:

cmsRun RelVal_HLT_GRun.py

To create a proper raw-data input file from a MC GEN-SIM file, in case the raw data file is no longer available somewhere, use cmsRun RelVal_DigiL1Raw_XXXX.py. Edit this config file to specify your Monte Carlo input file (e.g., from CASTOR, EOS, or locally on your PC).

For experts: The script "./getHLT.csh" creates the self-contained python cfg files called ../test/OnLine_HLT_XXXX_cfg.py, as well as the python cff files for use with cmsDriver ../python/HLT_XXX_cff.py, using ConfDB. The self-contained cfg files ../test/OnLine_HLT_XXXX_cfg.py must yield identical HLT results to the cmsDriver generated cfg files RelVal_HLT_XXXX.py. This tests whether the online and cmsDriver configurations match properly, as the fomer are used for real data and the latter for MC simulation.

Below you'll find the instructions for the open or anyhow recent CMSSW releases. Specific instructions for the older CMSSW releases and data taking scenarios can be found in SWGuideLegacyGlobalHLT.

Check for bug fixes for the L1 trigger here: SWGuideL1TriggerTags

Trigger development for Run-2

This part of the wiki describes how to develop new triggers and test them. In case you want your development integrated into the CMS trigger online, pay attention to the workflow achieving this goal, documented in detail here.

Specifically note that the current trigger development release is 92X. Thus, for any new C++ code needed for your development, you need to provide a pull request for 93X first and then a backport PR for 92X. For added module parameters and functionality, you must use either a fillDescriptions(.) method (preferred) or existsAs (deprecated) to take care of older py configs not yet containing the new parameters. The new parameters' default values must be set in such a way so that the old behaviour of the module is reproduced! Also, the result of the code must not depend on whether any parameter is present in the py config or not, only on its default value (in fillDescriptions) or its value in the py config! This allows us to integrate code developments independently (and earlier) than updating the HLT menu (python config) itself with your changes: the python update requires ConfDB parsing to discover the new parameters, followed by a migration of the HLT menu within ConfDB to the newly parsed code template.

Preparing a menu in ConfDB

The master development table for 2018 - with phase-1 geometry - is currently located in /dev/CMSSW_10_1_0/HLT, from which the following subtables are derived: /dev/CMSSW_10_1_0/GRun for pp, /dev/CMSSW_10_1_0/HIon for PbPb (currently dummy), /dev/CMSSW_10_1_0/PRef for ppRef5TeV, and /dev/CMSSW_10_1_0/PIon for pPb (currently dummy). Have a look at the recent developments of the CMSSW_9_0_0 Run-2 menu CMSSW_9_0_1 Run-2 menu CMSSW_9_2_0 Run-2 menu CMSSW_10_0_0 Run-2 menu CMSSW_10_1_0 Run-2 menu . For the development of modified or new paths, you should start from the most recent master table: copy it to your user area in ConfDB and use it as a starting point for your development. You may remove paths irrelevant for your task, but make sure you keep the HLTriggerFirstPath and HLTriggerFinalPath from the master or GRun table as the first and last path in your area, the latter just before any Endpath. The HLT menu is continuously developed, both in terms of physics and the technical infrastructure such as tracking or muon reconstruction. Thus, if you take too long time to develop, test and submit your request for integration, we'll ask you to re-implement your development based on the most recent master table. Make sure that your changes are compatible with the master table (ie, not affecting other paths unless explicitly required), not just the specific subtable you are aiming for!

New triggers must be fully configured and self-contained, and made available in ConfDB - ie, for integration into the master and GRun tables, we do not accept customisations on top of configurations in ConfDB. If plugins are missing in ConfDB, let us know the details asap. hltIntegration tests must (be rerun to) use those final configurations. Beware of our conventions (eg, module labels start with hlt*, sequences start with HLT*, etc.). Also provide stream/dataset assignments for your paths in the JIRA ticket. Many more details are summarised here

Preparing a 100X CMSSW developer area for 2017

The generic procedure is similar to creating a working area. Development for/in 2018 must use CMSSW_10_1_X, the release cycle used by HLT for the initial 2018 data taking period. You need to use the most recent CMSSW_10_1_X release.

Attention: the instructions posted in the following are to be used for 2018 triggers and their further development. If you want to run 2017 menus, you need to use a 92X release (80X for 2016 menus). Please see specific instructions for processing TSG MC samples below!

It is now possible to use the hltGetConfiguration cmdline parameter --l1-emulator 'Full/FullMC/FullSim/FullSimHcalTP' in order to directly process an older RAW file and run the L1T emulator on the fly together with the HLT (cmdline parameters --l1 ... and --l1Xml ... now also work). In case you still want to produce an explicitly repacked RAW file, use cmsDriver to run L1REPACK:Full/FullMC/FullSim/FullSimHcalTP with the option --era Run2_2017. For details on available L1REPACK options see here

Running the HLT with CMSSW_10_1_X on 92X, 100X TSG samples and 2017 data

To run on 2018 data, the latest GlobalTag is the following: 101X_dataRun2_HLT_v7

To run the HLT on 92X TSG samples, you need to:

• use the following global tag (see SpecialGlobalTagsForTSG):
  MC 92X: 101X_mc2017_realistic_TSG_2018_04_09_20_43_53
  MC 100X: 101X_upgrade2018_realistic_TSG_2018_04_16_20_31_16
  Data 2017 (92X): 101X_dataRun2_relval_Candidate_2018_03_28_20_20_48
  
  
• For 92X, 94X samples (both MC and data): apply the customization function "customiseFor2017DtUnpacking" in HLTrigger/Configuration/python/customizeHLTforCMSSW.py (see example below). It is needed to make the DT unpacker compatible with the old data format (see #21474 and CMSHLT-1709).

92X

100X

94X

For L1T prescales, see this info here

cmsrel CMSSW_10_1_11_patch1
cd CMSSW_10_1_11_patch1/src
cmsenv
git cms-addpkg HLTrigger/Configuration

# Dependencies and Compilation
git cms-checkdeps -A -a
scram b -j 6
rehash

# Runtime test
cd HLTrigger/Configuration/test

# running on 92X MC files with 101X:
hltGetConfiguration /dev/CMSSW_10_1_0/GRun \
  --globaltag 101X_mc2017_realistic_TSG_2018_04_09_20_43_53 \
  --path HLTriggerFirstPath,HLTriggerFinalPath,HLTAnalyzerEndpath \
  --input root://cms-xrd-global.cern.ch//store/mc/RunIISummer17DRStdmix/TT_TuneCUETP8M2T4_13TeV-powheg-pythia8/GEN-SIM-RAW/NZSFlatPU28to62_92X_upgrade2017_realistic_v10-v2/50000/0043237F-0DA1-E711-AD49-346AC29F11B8.root \
  --mc --process MYHLT --full --offline \
  --l1-emulator uGT \
  --l1 L1Menu_Collisions2018_v1_0_0-d1_xml \
  --customise HLTrigger/Configuration/customizeHLTforCMSSW.customiseFor2017DtUnpacking \
  --prescale none --max-events 10 --output none > hlt.py

cmsRun hlt.py >& logHLT &

In order to enable other L1 triggers and apply different L1 prescales, one can use git cms-addpkg L1Trigger /L1TGlobal and then edit the files L1Trigger/L1TGlobal/data/Luminosity/startup/mask-slim.xml and L1Trigger/L1TGlobal/data/Luminosity/startup/prescale-slim.xml.

Dumping the latest HLT menu configuration

BEFORE you start, read this section in FULL!

When you have extracted a py file, DO NOT make any changes to it (unless described below or you know what you are doing) else it invalidates any support from STORM.

Dumping a self-contained cfg file for direct use with cmsRun

The GRun/HIon/PRef/PIon configurations dumped into the HLTrigger/Configuration are always working in the corresponding release (the PIon and HIon tables are a simplistic placeholder menu). Development of these menus takes place in ConfDB. In case you live on the bleeding edge and want to try the most-recent integrated table from ConfDB, dump it with:

# Running on real data (reset HLT prescales all to 1)
hltGetConfiguration /dev/CMSSW_10_1_0/GRun --full --offline --data --unprescale --process TEST --globaltag auto:run2_hlt_GRun > hlt.py

or:

# Running on Monte Carlo simulation (reset HLT prescales all to 1)
hltGetConfiguration /dev/CMSSW_10_1_0/GRun --full --offline --mc --unprescale --process TEST --globaltag auto:run2_mc_GRun > hlt.py

Besides the --data versus --mc difference, note the different GlobalTag for processing real data versus Monte Carlo simulation. In case you want to process the special TSG samples, use the special GTs as described above, and add instructions do run the new L1 as described below.

To run your own menu containing only HLT paths (and eventually additional ESModules), but using the services and event setup from the master configuration, use (for real data):

hltGetConfiguration /your/test/menu/with/only/paths --full --offline --data --unprescale --process TEST --globaltag auto:run2_hlt_GRun --setup /dev/CMSSW_10_1_0/GRun > hlt.py

In case of Monte-Carlo simulations, replace --data by --mc and --globaltag auto:run2_hlt_GRun by --globaltag auto:run2_mc_GRun. In case you want to process the special TSG samples, use the special GTs as described above, and add instructions do run the new L1 as described in the following. Remember: HLTriggerFirstPath,HLTriggerFinalPath as first and last path, the latter just before any endpath, in your own area!

Adding and running the new L1 emulator and new L1 menu

This is only neded to process raw files made in earlier releases, such as the TSG 81X and 83X samples. Those releases emulated the old L1, so we need to add instructions to emulate the new L1:

Append --l1-emulator 'Full/FullMC' for real-data/MC to the hltGetConfiguration command, like this:

hltGetConfiguration /dev/CMSSW_10_1_0/GRun --full --offline --mc --unprescale --process TEST --globaltag auto:run2_mc_GRun --l1-emulator 'FullMC'  > hlt_stage1.py

For details on available L1REPACK options see here

In case you want to process the special TSG samples, use the special GTs as described above. Note: we no longer use xml files for new L1T menus. The L1T menu is set by the GT; in case your favorite GT does not contain the desired L1T menu, you can force it by adding --l1=L1Menu_Collisions2016... in the hltGetConfiguration command in case the menu is at least in CondDB.

Dumping a configuration-fragment cff file for use with cmsDriver.py

If your own ConfDB area contains ONLY paths, and at most a subset of (possibly modified) setup modules, with the rest to be taken from the standard setup, prepare the cff dump as follows:

hltGetConfiguration --cff --offline /your/test/menu/with/only/paths --setup /dev/CMSSW_10_1_0/GRun > HLT_User_cff.py

You need to have HLTriggerFirstPath,HLTriggerFinalPath as first and last path, the latter just before any endpath, in your own area!

If your area contains all setup modules required for your paths, prepare the cff dump in one go:

hltGetConfiguration --cff --offline /your/fully/self/contained/area  > HLT_User_cff.py

Remember: HLTriggerFirstPath,HLTriggerFinalPath as first and last path, the latter just before any endpath, in your own area!

Move the resulting file HLT_User_cff.py to src/HLTrigger/Configuration/python and run scram b to compile it.

Now you can use cmsDriver.py, specifying the HLT step as --step=HLT:User. Remember to specify the correct GlobalTag and any required Era customisation on the cmsDriver.py commandline, such as: --era=Run2_2016 or --era=Run2_HI,Run2_2016.

If you need to rerun L1, to get the new L1, use a gen-sim file as input and specify: --step DIGI,L1,DIGI2RAW,HLT:User.

Note: we no longer use xml files for new L1T menus. The L1T menu is set by the GT; in case your favorite GT does not contain the desired L1T menu, you can force it by adding --l1=L1Menu_Collisions2016... in the hltGetConfiguration command in case the menu is at least in CondDB..

Apply (and customise) L1 Prescales

Note: these instructions require running the L1 uGT emulator. In case one needs to run a wider L1 emulation, beyond just the uGT, one needs to check and apply also the latest L1 emulator instructions.

1) Setup

On top of the basic HLT setup described above, the following subdirectory is needed:

 L1Trigger/L1TGlobal/data/Luminosity/startup 

One should add the L1 menu/prescales/masks xml files in this subdirectory, and force their use by adding the "hackConditions" module by hand in the HLT config file.

For example, to use the 2018 L1 menu v1, the L1 prescales from run 317696 (taken on 11/06/2018), and trivial masks (all set to "1"):

git cms-addpkg L1Trigger/L1TGlobal
mkdir -p L1Trigger/L1TGlobal/data/Luminosity/startup

cp /afs/cern.ch/user/n/ndaci/public/STEAM/L1Menus/2018/L1Menu_Collisions2018_v1_0_0.xml L1Trigger/L1TGlobal/data/Luminosity/startup
cp /afs/cern.ch/user/n/ndaci/public/STEAM/Prescales/2018/L1PS_2018_06_13_Run317696.xml L1Trigger/L1TGlobal/data/Luminosity/startup
cp /afs/cern.ch/user/n/ndaci/public/STEAM/Prescales/2018/mask-trivial.xml L1Trigger/L1TGlobal/data/Luminosity/startup

2) Adjust the L1 PS of a given L1 algo

2.1) Identify the bit number of the algo of interest by checking the L1 menu xml file (or the WBM PrescaleSets webpage).

2.2) Open the L1 PS xml file: each row corresponds to a given algo; the first number of a row is the bit number of the algo; all other numbers are the L1 PS values in all columns.

2.3) Edit the line corresponding to the L1 algo of interest and set the desired values for each column.

3) Customize the HLT config file

3.1) Produce the HLT config file (hltGetConfiguration command) as described in the previous subsection (SWGuideGlobalHLT#Dumping_the_latest_HLT_menu_conf).

3.1.1) Specify the L1 menu xml file, as well as the HLT PS column of interest.

3.1.2) For example, to use the 2018 L1 menu xml file and the 2.0e34 column, specify the following options (among others) to the hltGetConfiguration command: "--prescale 2.0e34" and "--l1Xml L1Menu_Collisions2018_v1_0_0.xml"

3.1.3) Specify also the L1 uGT emulation requirement: "--l1-emulator uGT"

3.2) Edit the config file and add at the end the following lines:

process.load('L1Trigger.L1TGlobal.hackConditions_cff')
process.L1TGlobalPrescalesVetos.PrescaleXMLFile = cms.string('L1PS_2018_06_13_Run317696.xml')
process.L1TGlobalPrescalesVetos.FinOrMaskXMLFile = cms.string('mask-trivial.xml')
process.simGtStage2Digis.AlgorithmTriggersUnmasked = cms.bool(False)
process.simGtStage2Digis.AlgorithmTriggersUnprescaled = cms.bool(False)
process.simGtStage2Digis.PrescaleSet = cms.uint32(2)

Here, the "PrescaleSet" parameter refers to the column, and #2 is the 2.0e34 column. This parameter, and the name of the column you specify for the HLT PS through the "--prescale" option, must be chosen coherently.

Using an up-to-date GlobalTag

The GlobalTag currently in the menu is for online running. Offline, you should use a globaltag matching the data or MC samples being used. To first approximation, one can use a "symbolic" GlobalTag which is defined in Configuration/AlCa/python/autoCond.py (but please notice that such tags are useful for testing, but should never be used to make performance-related studies; for those studies you must use the specific GlobalTag's as listed at the top of this page):

For real data with 2017 geometry:

hltGetConfiguration ... --data --globaltag auto:run2_hlt_GRun ....

For Monte-Carlo simulation with 2017 geometry:

hltGetConfiguration ... --mc --globaltag auto:run2_mc_GRun ....

For real data with 2016 geometry:

hltGetConfiguration ... --data --globaltag auto:run2_hlt_GRun2016 ....

For Monte-Carlo simulation with 2016 geometry:

hltGetConfiguration ... --mc --globaltag auto:run2_mc_GRun2016 ....

If needed for your studies, you can also provide an explicit GlobalTag as parameter for --globaltag.

See above for the special GTs to be used with the special 81X or 83X TSG samples!

Caveat please note that, in general, there is no guarantee that using a different CMSSW release will reproduce the same HLT results. The global tag is just one of the things that may affect the results.

Testing your HLT paths

To test new paths before submitting them for integration, create an empty configuration with all the paths you are going to submit, and run the hltIntegrationTests script, either on real data:

hltIntegrationTests /your/test/menu/with/only/paths -s /dev/CMSSW_10_1_0/GRun -i some/input/data  -x "--globaltag auto:run2_hlt_GRun" 

or on MC:

hltIntegrationTests /your/test/menu/with/only/paths -s /dev/CMSSW_10_1_0/GRun -i some/input/data  --mc -x "--globaltag auto:run2_mc_GRun"

if you are using a different L1 menu, please add the hltGetConfiguration parameters using '-x':

... -x " --l1-emulator FullMC " -x " --l1 L1Menu_Collisions2017_dev_r5_m4_patch_xml "

(in case the L1 menu is available only as .xml file, copy it in L1Trigger /L1TGlobal/data/Luminosity/startup and use --l1Xml option)

Please, make sure that for all paths tested at least one event gets accepted in the test (you must check the log output for that).

The -s option is used to tell the script to take the setup (ESSources, ESProducers, Services) from the master table, so you don;t need to include them in your area. You do need to include any new module you may need.

The -i option is used to specify an input file, in any format understood by the PoolSource . It can be a local (file:...), a file on castor identified by its LFN (/store/...), or any other supported format.

The -x "--globaltag auto:hltonline" or -x "--globaltag auto:run2_mc_GRun" sets the globaltag for either real data or Monte-Carlo simulations. For the latter, also add --mc on the hltIntegrationTests command-line.

The comands to rerun the L1 emulator as used above in the hltGetConfiguration commands can be passed on by wrapping them in repeated -x "--..."= commands on the ==hltIntegrationTests cmdline, eg, -x "--l1-emulator 'stage1,gt'" -x "--l1Xml L1Menu _Collisions2015_25ns_v2_L1T_Scales_20141121_Imp0_0x1030.xml"

For more options, see hltIntegrationTests -h .

Submission for integration into CMSSW (CMSHLT-JIRA tickets)

Follow the instructions given here to get your paths integrateted in the HLT menu.

Legacy HLT menus and older releases

CMSSW_10_3_X (2018 HIon data taking release)

2018 Run-2 development HLT menus: GRun for pp (25ns), HIon for PbPb, PRef for pp5TeVref, and PIon for pPb.

setenv SCRAM_ARCH slc6_amd64_gcc700
cmsrel CMSSW_10_3_1_patch3
cd CMSSW_10_3_1_patch3/src
cmsenv

git cms-addpkg HLTrigger/Configuration

scram build -j 4
cd HLTrigger/Configuration/test

CMSSW_10_1_X (2018 pp data-taking release)

2018 Run-2 development HLT menus: GRun for pp (25ns), HIon for PbPb, PRef for pp5TeVref, and PIon for pPb.

setenv SCRAM_ARCH slc6_amd64_gcc630
cmsrel CMSSW_10_1_11_patch1
cd CMSSW_10_1_11_patch1/src
cmsenv

git cms-addpkg HLTrigger/Configuration

scram build -j 4
cd HLTrigger/Configuration/test

CMSSW_9_2_X (2017 collision-data-taking release)

2017 Run-2 development HLT menus: GRun for pp (25ns), HIon for PbPb, PRef for pp5TeVref, and PIon for pPb.

setenv SCRAM_ARCH slc6_amd64_gcc530
cmsrel CMSSW_9_2_15
cd CMSSW_9_2_15/src
cmsenv

git cms-addpkg HLTrigger/Configuration

scram build -j 4
cd HLTrigger/Configuration/test

CMSSW_8_0_X (2016 pp and pPb collision-data-taking release for HLT)

2016 Run-2 development HLT menus: GRun for pp (25ns), HIon for PbPb, PRef for pp5TeVref, and PIon for pPb.

setenv SCRAM_ARCH slc6_amd64_gcc530
cmsrel CMSSW_8_0_31_patch1
cd CMSSW_8_0_31_patch1/src
cmsenv

git cms-addpkg HLTrigger/Configuration

scram build -j 4
cd HLTrigger/Configuration/test

CMSSW_7_5_X (2015 PbPb collision-data-taking release for HLT)

2015 Run-2 HLT menus

setenv SCRAM_ARCH slc6_amd64_gcc491
cmsrel CMSSW_7_5_9
cd CMSSW_7_5_9/src
cmsenv

git cms-addpkg HLTrigger/Configuration

scram build -j 4
cd HLTrigger/Configuration/test

CMSSW_7_4_X (2015 pp collision-data-taking release for HLT)

2015 Run-2 HLT menus

setenv SCRAM_ARCH slc6_amd64_gcc491
cmsrel CMSSW_7_4_16_patch2
cd CMSSW_7_4_16_patch2/src
cmsenv

git cms-addpkg HLTrigger/Configuration
git cms-merge-topic cms-tsg-storm:74XHLTppRefMenu # for the ppRef5TeV menu

scram build -j 4
cd HLTrigger/Configuration/test

CMSSW_5_3_X (Legacy release for Run-I data taking period - 2011/2012/2013)

2013 PIon, 2012 GRun, 2011 HIon, and 2011 resurrected HLT menus

setenv SCRAM_ARCH slc6_amd64_gcc472
cmsrel CMSSW_5_3_38
cd CMSSW_5_3_38/src
cmsenv

git cms-addpkg HLTrigger/Configuration

scram build -j 4
cd HLTrigger/Configuration/test

Useful links

https://twiki.cern.ch/twiki/bin/viewauth/CMS/SpecialGlobalTagsForTSG

https://twiki.cern.ch/twiki/bin/viewauth/CMS/TriggerTrainModel

https://twiki.cern.ch/twiki/bin/view/CMSPublic/SWGuideL1TStage2Instructions

MC samples 2018 (special TSG samples) google doc

HLT menu google doc

https://twiki.cern.ch/twiki/bin/view/CMS/HLTinCMSSW101X

https://twiki.cern.ch/twiki/bin/view/CMS/ConfDB1000

https://twiki.cern.ch/twiki/bin/view/CMS/ConfDB1010

Special data for rate studies: https://twiki.cern.ch/twiki/bin/view/CMS/TriggerStudiesChangesInDataTaking2018#High_rate_ephemeral_datasets_for

Random test files on EOS

92X MC

root://eoscms.cern.ch//eos/cms/store/mc/RunIISummer17DRStdmix/TT_TuneCUETP8M2T4_13TeV-powheg-pythia8/GEN-SIM-RAW/NZSFlatPU28to62_92X_upgrade2017_realistic_v10-v2/50000/8CAC0136-BFA0-E711-82A8-0025904B2C4C.root

2017 data

root://eoscms.cern.ch//eos/cms/store/data/Run2017C/MuonEG/RAW/v1/000/302/029/00000/F4E8E245-608D-E711-9483-02163E014146.root

Responsible: MartinGrunewald