TkDQM: Issue with Pixel Phase1 Online DQM

[sroychow]

Problem reported in 2022 w.r.t Online DQM pixel client - LS missing from processed plots. The problem seems to have appeared this year as well. Please review this as soon as possible. A set of runs will be provided by online experts.

[alaperto]

Hello, this is the DQM package that builds the Pixel Summary Maps.

Essentially there are 2 files:

• .cc file
• .py file

---

Inside the .cc file, there are separate actions for the Online and Offline Summary, based on this Bool
if (runOnEndLumi_) --> Online
if (!runOnEndLumi_) --> Offline

---

The Offline Summary map (also called "Grand" Summary) is defined here: 12x4

The Online Summary map is defined here: 2x4

---

Both Online and Offline have 5 Report maps, while Cosmics only on 3.

RunQTests_online = cms.Sequence(ADCQTester * NumClustersQTester * NumDigisQTester * SizeQTester * ChargeQTester)
RunQTests_offline = cms.Sequence(ADCQTester_offline * NumClustersQTester_offline * NumDigisQTester_offline * SizeQTester_offline * ChargeQTester_offline)
RunQTests_cosmics = cms.Sequence(NumDigisQTester_cosmics * SizeQTester_cosmics * ChargeQTester_cosmics)

The Report maps are defined here: 12x4

• the 2D histogram is 12x4, but only 40 bins (4 Bxy * 4 layers + 8 HC_xy_Z * 3 disks) are filled. The X axis corresponds to: BmO, BmI, BpO, BpI, HCmO_1, HCmO_2, HCmI_1, HCmI_2, HCpO_1, HCpO_2, HCpI_1, HCpI_2
• another Report map is build "new 6 bin ROC summary" with only 6 bins (4 layers + 2 rings): 2x4

I have not found in the code the threshold values for the 5 Report maps... Are they stored somewhere else?

The Report maps are stored in a sub-folder: PixelPhase1/Phase1_MechanicalView/

Example of Report maps for:

• Online
• Offline
• Offline Cosmics

Together with the 5 Report maps, there is also the "deadROCSummary" map in the Offline GUI, but it seems this ROC map (2x4) is not used in the Offline Summary (12x4).

The Offline Summary map is filled here, as the worst value of each bin:

• for each of the 40 bins, the "Grand" Summary is built as the worst value of all the 5 Report maps
• every block is set to 1 (100% good)
• if the block in one of the Report maps is worse than the current value, it is taken as the new current value
• the variable reportSummary (probably the number shown in the Report Workspace) is the average of the 40 "Grand" Summary bins

---

It seems that the Online Summary map (2x4) is filled only related to the "deadROCSummary" (2x4):

• for each of the 6 bins, if the value is < WarnThresholds (=0.1), this value is used in the Summary map
• else, if the value is > WarnThreshold but < Threshold (=0.2), 0.8 value is used in the Summary map
• else, 0 is used

deadRocThreshold and WarnThreshold are taken from the .py file:

  deadRocThresholds_ = conf_.getParameter<std::vector<double> >("DeadROCErrorThreshold");
  deadRocWarnThresholds_ = conf_.getParameter<std::vector<double> >("DeadROCWarningThreshold");

Here the 6 values for DeadROC Error/Warning Thresholds in the .py file.

    # Number of dead ROCs required to generate an error. Order must be layers 1-4, ring1, ring2.
    DeadROCErrorThreshold = cms.vdouble(0.2,0.2,0.2,0.2,0.2,0.2),
    DeadROCWarningThreshold = cms.vdouble(0.1,0.1,0.1,0.1,0.1,0.1)

There are always 6 values for Online/Offline/Cosmics, but maybe this is only used in Online since it is only referred to the 4 Layers + 2 Rings = 6 bins of the Online maps.