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:Ê The menus presented by S. Tapprogge during the last T/DAQ week will be assumed to be the baseline menus. From the presentation it is concluded that the low luminosity scenario now corresponds to a luminosity of 2 x 10^33 and that there is no longer a need for an intermediate luminosity scenario. This leads to three menus: two for low luminosity (one with and one without deferrals) and one for highÊluminosity. All menus do contain an item labeled with "Other" and corresponding to a LVL1 rate of 5 kHz. As this item is unspecified it is will not be taken into account for the time being. M. Abolins will generate new results (available here: low luminosity without deferrals, low luminosity with deferrals, high luminosity).Ê The "mini-scan" of the inner detector has still to be implemented in the paper model program, this will be done on a short time scale.Ê
2. ROS modelling: R. Cranfield continued with the implementation of a model of the ROS, using P.Golonka's code for the IOM model. Support for ROBIns was included (in the model ROBIns immediately respond to requests).
Testing of the model will start now. The note concerning parameters to be
measured need to be updated somewhat and to be included in the document from
P. Golonka and K. Korcyl, as discussed in the previous meeting.
ÊThe short note on "hot spots" by K.Korcyl, discussed in the previous meeting
as part of the discussion on the ROS model, will be available on a short
time scale.
3. Data Collection modelling and HLT modelling: A new draft document on the DC model parameters is available. Calibration measurements for various components are going on. The models and parameters for the SFI and DFM are now available, a simulation of the test systems used for obtaining the parameters is also available. The next step consists of checking simulation against measurement results. The message passing model has been extended with an implementation of flow control and of interrupt handling, also to be tested soon. Handling of multiple interfaces is needed for the SFI, but also for the ROS, the same solution is to be used in both cases. R. Cranfield will wait for P. Golonka to finish cleaning up of the code. Internally in the ROS messages have to be used not needed elsewhere in the system. The names of these messages are to be prefixed with an identifier indicating their special type. Parameters for the supervisor and L2PU are still missing.Ê
4. at2sim: A major cleanup of the makefiles has taken place. The
whole program is now statically linked with as result that theÊ startup time
is much shorter than before (now < 1/2 s). The new version has still to be checked into the repository.
5. Simdaq: no news.
6. Chiba City: no news.
7. Workplan: From the minutes of the previous meeting: "by the beginning of January the new version (of at2sim) should be functional. Then by the end of January the final parameters should be available and the results of a 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 analyzed, which should be done in time for the March T/DAQ workshop." Conclusion: the work is proceeding more or less as planned.
8. Next meeting: Wednesday 29 January 2003, 15.30 h CERN time.
9. AOB:Ê D. Francis mentioned the missing parameter results and also that the configuration of the small scale testbed system needs to be defined. This is to be discussed in the Data-Flow meeting of 13 January,
J. Vermeulen, 13 January 2002
Ê