Task Name : Review of parameters and update of paper model
Task ID   : Modelling-1-PaperModelUpdate
Organiser : A. Amadon and J. Vermeulen
Need      : Essential
Aims / Objectives : specification of parameters relevant for the LVL2 system
            and paper model study of the LVL2 system
Definition/Approach : Paper study, spreadsheet
Duration : until December 1998
Deliverables : Spreadsheet conforming to current trigger/buffer/LVL1 &
               ATLAS trigger architecture as desribed in and trigger menus as specified
               in forthcoming documents
Milestones :
   internal: October 1998, specfication and spreadsheet available
   external: October 1998, specfication available
             December 1998, spreadsheet available
Dependencies : input from detector and physics groups
Partners : NIKHEF, SACLAY, Michigan State University
Resources :
Risks : medium (for internal deadline)
Status / Date : June 7 1998 : in progress, internal and external milestones had to
                be shifted to October and December 1998 respectively due to the
                necessary input not being available.
Comment :

Task Name : Implementation and study of behaviour of the full but generic
            system model
Task ID   : Modelling-2-FullGenericModel
Organiser : Modelling convenors
Need      : Essential
Aims / Objectives : study different options with respect to possible designs
                     of the LVL2 system
Definition/Approach : discrete event simulation in C++ with SIMDAQ-C++
Duration : until October 1999
Deliverables :
        o Discrete event simulation program(s) able to model the various
          options under study
        o Document(s) describing the model results for each with one or
          more parameter choices
Milestones :
   internal: October1998   : models for sequential processing, models for
                             "model C processors" and SFIs, results pull model B, RSIs,
                             additional models and refinements (e.g. model for shared
                             bus, finite buffer sizes)
             November 1998 : results from models for sequential processing
             December 1998 : All components of full system, also used in the
                             paper model, implemented and tested
   external: June 1999 : models and results available
Dependencies : Modelling-1-PaperModelUpdate
Partners : ANL, CRACOW, NIKHEF, SACLAY, UCL
Resources :
Risks : medium
Status / Date : June 7 1998 : in progress, tasks for internal milestone of June 1998 only
                partially finished (implementation pull model B and of RSIs),
                milestones updated.
Comment : The implementation of the pull model option of model B and of the RSIs has
          solved a number of issues that are also important for the implementation of
          model C

Task Name : Implementation and study of behaviour of generic models (i.e.
            without detailed models of different technologies used) models
            of test systems
Task ID   : Modelling-3-GenericTestSystemModels
Organiser : Modelling convenors
Need      :
Aims / Objectives : verification of correctness of simulation and help in
                    understanding experimental results
Definition/Approach : discrete event simulation in C++ with SIMDAQ-C++ or /
             and using Ptolemy
Duration : until October 1999
Deliverables :
        o Simulation programs to model test setups with small component counts
        o Documents comparing above programs to measured test setup results
Milestones :
   internal: October 1998 : DS-link model B : model and results
   external: October 1999 : models and results
Dependencies : Documents describing setup and measurement results of test
systems
Partners : ANL, NIKHEF
Resources :
Risks : medium
Status / Date : June 7 1998 : in progress. Task for internal milestone of June
                1998 not completely finished, due to unforeseen difficulties with
                task 2 (implementation of pull model B and of RSIs) and due to
                unforeseen requirement with respect to required modifications of
                Simdaq-C++. Milestone shifted to October.
Comment : will develop, more partners to be added later. ANL (B. Blair) will
          attempt to develop a generic model for ATM based demonstrators.

Task Name : Implementation and study of behaviour of full models (i.e. with
            detailed models of different technologies used) of test systems
Task ID   : Modelling-4-TechnologyModels
Organiser : Modelling convenors
Need      : Driven by test systems
Aims / Objectives : verification of correctness of simulation and help in
                    understanding experimental results
Definition/Approach : discrete event simulation in C++ with SIMDAQ-C++ or /
             and using Ptolemy
Duration : until October 1999
Deliverables :
        o Refined simulation programs with tuned parameters for test setups
        o Document(s) making comparisons with measured test setup performance
Milestones :
   internal:
   external: October 1999 : models and results
Dependencies : Can only follow test setups
Partners : ANL, CRACOW, MANCHESTER, NBI, SACLAY, RHUL
Resources :
Risks :
Status / Date :  June 7 1998 : to be planned before end of year
Comment : Previous work on ATM, SCI, DSlink models, etc. potentially useful
          for this task. Saclay (A. Amadon) will include suitably modified C++ code simulating
          ATM developed previously at Saclay in Simdaq-C++

Task Name : Refinement and study of behaviour of full models, with
            detailed models of different technologies, of the LVL2
            system, as needed
Task ID   : Modelling-5-FullModel
Organiser : Modelling convenors
Need      :
Aims / Objectives :
Definition/Approach :
Duration : until October 1999
Deliverables :
        o Refined full system simulation program with tuned parameters
        o Document describing expected system performance given `standard'
          menu choices
Milestones :
Dependencies : Modelling-2-FullGenericModel, Modelling-4-TechnologyModels
Partners :
Resources :
Risks : low
Status / Date : to be planned before end of year
Comment :