Skip to content

HLT data scouting in athenaMT

Rafal Bielski requested to merge rbielski/athena:hlt-data-scouting into master

This MR includes what's needed to configure and use basic data-scouting chains in HLT in athenaMT:

1. Small modifications to TriggerEDMSerialiserTool

Change the CollectionsToSerialize property from vector<string> to map<string,vector<uint16_t>>, with the meaning of mapping collection name to the vector of moduleIDs (HLT result types) where it should go. Propagate the moduleIdVec information through the configuration in initialize(). In fill(), add the serialised collection to all results where it should go. Extend the TriggerEDMSerialiserTool python class to make adding collections easier.

2. Refactor the PEB test

Isolate a common part between PEB and DS tests and put it into a separate python file. Fix the EDM serialisation in the PEB test (correct the collection list).

3. Add new test to provide an example on how data scouting should be configured

New job options and test with reference in TrigUpgradeTest. Four chains are configured, which give example on various DS/PEB combinations.

See also ATR-19360 and slides 4-7 in this presentation:
https://indico.cern.ch/event/798089/contributions/3316566/subcontributions/274992/attachments/1796828/2929570/20180215_Rafal.pdf

In this MR I only focused on the "offline" part, i.e. what sits under the Trigger directory and can be executed with offline athena. I will test and adapt the online part as the next step.

Edited by Rafal Bielski

Merge request reports