Mark
Javier, Yee, Paul
GARR
Loukik
Robert Stoy, DFN




Javier: Context:
We are talking about network performance monitoring.
This is related to EGEE DJRA4.2, and MJRA4.3
Initially we expected to have the model first, but then we'd have the prototype,
but Peter suggested that it would be good to have the two at the same time, to make sure everything works together.

So we start with our requirements from users, operations and middleware, and now we have to think about how we provide these things.

So UCL will take the technical lead creating the prototype, and Mark will try to provide as much input as possible.


Mark:
Try to define exactly what we want to do, and where we want to go.



Yee: We do need to find out where and how data will be stored, so we can effectively make a frontend.

We were thinking about writing a set of intermediate libraries to convert any monitoring system into a compliant one.
We can use this to discover what is important and what isn't.



Can we write down definitions of all the words.
What does strawman mean? It has no boundaries...
If someone writes down the boundaries of this problem, then things will be easier to understand.


Can Yee and Paul elaborate

Gloria: Can I write an example survey of required features from a measurement architecture, for the EDG stuff.


Javier: by the end of the year, would like a service that makes use of one of the architectures answering the queries we think we want, and also interfacing with R-GMA.

Timescales: the prototype can go ahead, starting from these questions. The survey needs to be finished by the end of October.

The survey needs to go on until the end of October, but if we can start making a prototype with something.
The product choice needs to be made by October, so we can think about an implementation.