DUNE DAQ/SC Meeting

US/Central
Description
    • 08:30 08:40
      Consortium News 10m
      Speakers: Alessandro Thea (STFC Rutherford Appleton Laboratory), Giovanna Lehmann (CERN)

      Minutes from CT Meeting September 28th 2020

      Present:Giovanna, Georgia, Kurt, Phil, Giles, Alec, Bonnie, Pengfei

      Excused: Josh, Roland, Simon, Jim, Dave, Alessandro

      1. General news:
        1. Discussions with ND ongoing on how far to push synergies in the DAQ/SC areas: need to pay attention that there is no net loss of manpower for FD;
        2. Feedback received from collaborators on possible improvements of DAQ/SC consortium functioning; preparing a proposal to take it into account;
        3. Action on all coordinators: check schedule with Dave and make a last adjustment before release (after release the schedule will be frozen for several months and tracking of progress will be done regularly at CT meetings).
      2. Data Selection:
        1. Preparing meeting to discuss implementation of applications within appfwk;
        2. Preparing discussion on metrics to use to evaluate algorithms;
        3. Check against schedule to be done;
        4. At collaboration week: discussed how to support LE physics, in particular explored how TPs may be used and how a higher rate of localised triggers (60 cm for blips, 1 APA+adjacent for neutrons).
      3. MiniDAQ:
        1. Setup working at Columbia;
        2. SW installation done using basic readout application;
        3. Switching as soon as possible to readout application based on appfwk (FLX readout, sampled local trigger, data storage);
        4. Once this setup is fully debugged, packaged and documented it may be applied to ICEBERG.
      4. Dataflow:
        1. Message passing API almost final, with first implementation based on ZMQ (to be presented at DF meeting); Aim at presenting API at Monday's general DAQ/SC meeting;
        2. Data dispatcher: completed first sprint which achieved its goals; second sprint being prepared aiming at more sophisticated format; after second sprint it may be possible to integrate the data writing with the readout application for minidaq.
        3. At collaboration week: a presentation was made on the requirements capture for the analysis framework; no obvious input regarding data format to take home for DAQ -> we continue with the planned developments.
      5. Control, configuration and monitoring:
        1. Work to include configuration features inside appfwk well advanced; will aim at merging with main branch of appfwk this or next week
        2. Utility to send commands (with data payloads) based on a REST implementation almost ready;
        3. Started work on operational monitoring: technology evaluation for data distribution & archival, started defining API to allow user code to publish data and/or to subscribe to data. Will be presented at CCM meeting next week and thereafter at general meeting.
      6. Facility:
        1. ICEBERG support: data acquisition not yet fully integrated with electronics but almost ready. Based on protodune DAQ;
        2. At collaboration week:
          • computing/networking session in which use cases of WAN link utilisation were discussed; 30 PB/year to storage; peak throughput of 100 Gbps for SNB data transfer; headroom between average and peak bandwidth to be used ad hoc for additional data transfers to be processed at FNAL; high availability requirement (10G) for all control aspects: SURF should always be reachable in order to monitor and control the status of the experiment.
          • Presentation on HW database: to be explored asap how it fits DAQ/SC needs in order to learn how to use it and provide early feedback.
      7. SW Coordination:
        1. Progress on several aspects, from documentation organisation to automatic building, from multi-package development environment to DAQ release manifest file;
        2. Action: Prepare a clear over view of the roadmap such that CT members have a better understanding of what features will become available when.
      8. Agenda for next DAQ/SC meeting:
        1. API of message passing
        2. Experience of implementing the readout application within the appfwk
    • 08:40 09:00
      Experience of implementing the readout application within the appfwk 20m
      Speaker: Roland Sipos (CERN)