Argonne: R.Blair (part-time)
UCL: R.Cranfield
Michigan: M. Abolins
NIKHEF : J.Vermeulen
CERN: D. Francis, P.Golonka
Cracow: K.Korcyl
1. Paper modelling: M. Abolins has had no succes in obtaining more detailed information on the trigger menus described by V. Vercesi. The results for the trigger menus as they are specified (i.e.with insufficient breakdown of jet trigger items) will be communicated to V. Vercesi. M. Abolins also mentioned that J. Baines is still interested in the "mini-scan" of the inner detector, J. Vermeulen will implement support for this in the paper model program.
2. ROS modelling: A list of parameters to be determined from measurements has been sent to the ROS group. Amendments on the basis of comments by G. Crone have to be made, a revised version of the document will be circulated and should be included in the overall document on parameters. Before the meeting R. Cranfield also circulated a short note on the implementation of a model for the "ROS modular unit" in at2sim. Work on implementing the model has started and should be completed by the end of the year. K. Korcyl asked about the way of multiplexing inside the "ROS modular unit". This could be either with point-to-point connections or a shared bus. R. Cranfield remarked that it is not clear whether the ROBIns need to be included in the model for studies of the full system. This gave rise to a discussion on the aim of modelling. Various opinions were presented. There seemed to be agreement with respect to the need of defining the level of detail of the models such that on one hand the amount of effort in implementing and working with the models is manageable and on the other hand the results obtained are useable. A criterion for usability is the possibility to identify bottle-necks for data transport or other potential problem areas. The paper model is a first step in this direction. With the present knowledge of switch behaviour and with new results becoming available on a short time scale from the testbed a step further is possible in the direction of identification of technology related factors. Already with the "old" version of at2sim (containing models for up-to-date switch technologies) critical combinations of traffic patterns and traffic intensity can be identified. With the help of the paper model can then be checked whether there is a chance for these combinations to occur. K. Korcyl will look into this, beginning of January a short note will be available.
3. Data Collection modelling and HLT modelling: P. Golonka and K. Korcyl have updated the draft document on the DC model parameters. Calibration measurements for various components are going on.
4. at2sim: Work is going on on the "new" at2sim, most aspects are stable, but models of nodes and scripts for describing various configurations are still missing. By the beginning of January the new version should be functional. Then by the end of January the final parameters should be available and the results of small scale testbed model should have been verified. By the end of February the full model, including LVL2 sequential processing and correct mapping of the ROBs should be running, so that a comparison with paper model results is possible. Next results for the full model have to be produced and analysed, which should be done in time for the March T/DAQ workshop.
5. Simdaq: no news.
6. Chiba City: no news.
7. Workplan: See item 4. By the end of April tbackup documents on parameters, architecture, at2sim model, should be available
8. Next meeting: Tuesday 7 January 2003, 15.30 h CERN time.
9. AOB: The presentation in the December ATLAS T/DAQ week were discussed. M. Abolins will take care of the presentation on parameters, P. Golonka of the presentation on modelling status. The directions outlined in a mail from D. Francis will be respected. J. Vermeulen will prepare initial versions of the presentations.
J. Vermeulen, 5 December 2002