Architecture Working Group Meeting. 22-Nov-2002

											GMO
										     22-Nov-2002 


The purpose of the meeting was to agree on the mandate and organise the work in view of the TDAQ week.

 A common agreement was reached on the fact that we are asked to:

   1) focus on the two questions related to the TDAQ architecture with respect to deferrals:

		- Multiplexing of the detector read-out
		- growing the system size

      multiplexing may be done between the ROD and the ROB and/or between the ROB and the ROS
      the system may be grown (scaled) in size in a flat or a sliced manner

   2) Catalog the architectural variants related to multiplexing and scaling.

   3) Variants are discussed from the logical (or architectural) point of view,
      with as less reference to implementation as it is possible.

   4) Understand the implications which architectural variants may have on the
      overall system and on sub-systems (e.g. HLT). Understand in particular if
      any variants have effects on the system such that they should be excluded
      a priori (i.e. before embarking into more in depth studies).

   5) List what are the issues to be studied to technically (as opposed to cost-wise)
      evaluate the different variants. In particular individuate what issues are
      not covered by existing sub-system workplans. Propose a prioritization of
      these issues.

 Some initial background material:

	- the slides distributed by the TDMT ahead of the open meeting on 21-Nov.
	- the slides presented by David on Nov 21st.

 The different variants were briefly discussed with the aim to clearly define
 them. 

 It was agreed that a written descrition of the variants (point 2) will be circulated
 next monday and we should meet next week to start addressing points 4) and 5). 

 It is proposed to meet on

			Thursday 28-11 at 16:00