Argonne : R. Blair
CERN : P. Golonka, K. Korcyl, S. Wheeler
Michigan : M. Abolins
NIKHEF : J. Vermeulen, R. Scholte
1. Paper modelling: Computer model (simdaq) results for the new parameters, configuration and menus were obtained and compared to the paper model results. Some problems were found, giving rise to minor changes in the paper model results. An updated version of the document on paper model results is to be submitted as DAQ note.
2. ROS modelling: R. Cranfield has continued with the UML description. Diagrams and a document containing an overview of the diagrams are available. Implementation of the models can start, R. Cranfield intends to begin with the IOMs and then to incoporate the models of B. Rensch.
3. Data Collection modelling: P. Golonka showed preliminary results of a study on the scalability of the TCP/IP protocol (text, slides).
4. HLT modelling: S. Wheeler has rewritten the Ptolemy Event Builder model and made other modifications to at2sim, see the slides.
5. Ptolemy: K. Korcyl has finished the implementation of sequential processing in at2sim and checked it. A bug that gave rise to a core dump was fixed. Automatic selection of the switch configuration on the basis of the switch name has been included. Support for multi- and bradcasting is to be implemented in September.
6. Simdaq: The program has been modified so that it can handle the updated configuration and it has been run for this configuration and the new parameters and menus, as used for the paper model. A few distributions of the LVL2 decision time were shown (.pdf file). The results have been used for cross-checking the paper model results (see 1.)
7. "Chiba City": R. Blair reported that the benchmark programs of R. Hughes-Jones will be used to develop a better understanding of the behaviour of the system. There is no news from M. Brezuleanu on the model of the system.
8. Workplan and status in relation to workplan: An updated plan is not yet available, R. Blair takes care of updating it.
9. Next meeting: The next meeting (also via video) is foreseen for Wednesday 17 October, 15.30 h CERN time.
10. AOB: J. Vermeulen has asked R. Cranfield to take over his
role as modeling representative in the "Connect Forum" (QA group). R. Cranfield
has agreed under the condition that the amount of effort involved is relatively
small.
J. Vermeulen, 14 September 2001