Meeting Wed 6 August 2008

Participants: Julia Andreeva, Max Boehm, James Casey, Elisa Lanciotti.

Elisa had prepared some slides, see the attachement below.

The general conclusion from the meeting is that, before starting to develop any concrete prototype, the data model has to be defined in a more clear way.

More in detail:

• Elisa should provide a description of the schema using an ER diagram. It is not clear which tool to use to draw the diagram. Asked to Miguel Anjo, and waiting for a response.
• We should study more in detail what exactely the monitoring tools can provide us. And in which unit. Julia suggested to interact with Irina about metrics concerning the job processing for CMS, with Ricardo for ATLAS data transfer, and with Pablo for data transfer for CMS. This has to be done with high priority.
• The measurement should refer to a time period, rather than to a determined moment. This is a minor change in the tables. Already done.
• The structure of the gridmap should be different: there should be one gridmap per site. Then the group level should be the Activities. And for each group the inner rectangles should be the VO's. In this way it makes sense to use the gridmap, so that the size of the rectangle is proportional to the weight of each VO. This weight can be defined as the number of jobs (for example) or with other metrics. Or maybe the user could choose which parameter to use to set the size of the rectangles.

-- ElisaLanciotti - 08 Aug 2008