Folks, On 17 Jan 2005, at 19:24, Andrew Grimshaw wrote:
I think it is premature to think about a profile except in general terms of what we think should go in it. I think we need to think about spec’s, forming working groups to generate those specs, and so on. (Forming working groups involves charters, BOF’s, etc.)
Agreed. It is definitely too early to talk about profiling EMS (even candidate profiles). But, we really do need a WG started to define a minimal set of interfaces that actually allow us to run "hello World" on a Grid. I'm not sure I agree with, or really understand, Andrew's terms below, but then that is what a WG is all about.
In particular I think we should focus on the basic services we have identified so far (or as they were renamed to avoid confusion):
Service containers Actors Solutions Solution Factories Actor descriptors – which may be the same as something another group is working on – a description of the execution component. Hiro help me here.
And so on. We then need to make sure all of the players who may want to be included are contracted: e.g., Globus, EGEE, OMII, SGE, Platform, PBS, Virginia, and who ever else wants to be involved.
I would add to this list Unicore, NAREGI, and China Grid all of which have running interfaces parallel to the GRAM interface in Globus. All these groups have suggested to me in the past few days that they want to join a WG on this topic. OMII is discussing it actively. JSDL is ready for a delivery mechanism, .... In short, a BoF is in order. -- Take care: Dr. David Snelling < David . Snelling . UK . Fujitsu . com > Fujitsu Laboratories of Europe Hayes Park Central Hayes End Road Hayes, Middlesex UB4 8FE +44-208-606-4649 (Office) +44-208-606-4539 (Fax) +44-7768-807526 (Mobile)