FYI, attached are the minutes of the OGSA-WG telecon. Can I draw your attention to the Job Submission (or whatever its name ends up as) BoF being proposed. Cheers, Ali -- ---------------------------------------------------- |epcc| - Ali Anjomshoaa EPCC, University of Edinburgh James Clerk Maxwell Building Mayfield Road E-mail: ali@epcc.ed.ac.uk Edinburgh EH9 3JZ Phone: + 44 (0) 131 651 3388 United Kingdom Fax: + 44 (0) 131 650 6555 ------------------------------------------------------------- ---------- Forwarded message ---------- Date: Tue, 08 Feb 2005 16:32:16 +0900 From: Andreas Savva <andreas.savva@jp.fujitsu.com> To: 'ogsa-wg' <ogsa-wg@gridforum.org> Subject: [ogsa-wg] Teleconference minutes - February 7, 2005 Minutes attached. Gridforge: https://forge.gridforum.org/projects/ogsa-wg/document/minutes-20050207/en/1 or http://tinyurl.com/5fxfd/minutes-20050207/en/1 [Note that there is no call planned for Wednesday and next week is the face to face.] -- Andreas Savva Fujitsu Laboratories Ltd OGSA teleconference - 7 February 2005 ====================================== * Participants Jay Unger (IBM) Andreas Savva (Fujitsu) Dave Snelling (Fujitsu) Hrabri Rajic (Intel) Mark Morgan (UVa) Tom Maguire (IBM) Hiro Kishimoto (Fujitsu) Andrew Grimshaw (UVa) Dave Berry (NeSC) Minutes: Andreas Savva * Minutes for January 31 approved with no changes ** BoF preparation ** Simple Job Submission BoF - Driven by the need to define normatively a minimal subset of EM so that an EM pofile might go ahead. - Andrew is pulling things together - He has talked to Bill Nitzberg (area director) - He has told Stacey that a BoF application will be sent by Friday, the deadline for BoF scheduling requests. - Working on the draft application - Simple Job Submission may not be the right title. (It is more along the lines of 'sub-EM' or 'minimal EM.') - The proposed schedule is likely to be quite aggresive in order to meet various dependencies (and GGF expectations). For example, - First draft by June; and end of year as target delivery date - Clarified that there are 3 relevant documents: 1. BoF request (required to schedule a BoF) 2. Draft charter (good to have but not essential to schedule a BoF) 3. Starting point draft (as (2)) - For Friday's submission (1) and possibly part of (2) is good enough. - There is a minimal EM draft. (Based on the refactoring/renaming that Andrew, Mark, Chris did last year to address OGSA version 1.0 terminology problems.) - Starting with the EMS draft, ignore all services except container. Require a description of what to start and where (which container to use); and start it up. Return an EPR (or perhaps a Name). - Andrew and Mark worked with other UK participants in the mini-EMS F2F on this draft last week while in the UK. The result is a further simplification: http://www-unix.gridforum.org/mail_archive/ogsa-wg/2005/02/msg00009.html - It looks like there is a lot of interest. Many names were mentioned as possible chairs for the group. (No decisions yet.) - Agreed that when sending out the announcement all relevant people should also be cc-ed. - Need to also try to get Globus, SUN, and other parties involved. ** ByteIO BoF - Similar approach to the one described above. BoF request by Friday, tell the relevant area directors and so on. - Dave (B) will do the BoF application. - Again the BoF proposal was discussed during last week's UK meeting and many names came up as likely chairs. (Nothing fixed yet.) * UK meeting updates [The minutes of these meetings are being prepared by Mark and will be distributed once the participants review them. Very brief notes here.] ** OGSA Data mini-F2F - Dave (B) and Andrew Grimshaw gave an update. - Things seem to be coming together for a general data architecture - Data transport as an issue left for discussion at GGF13 ** OGSA EMS mini-F2F - Two day meeting with a number of local participants - Worked on simplifying the EMS draft. Produced the draft sent to the list (see url above). Some UML included. (Note that people listed have not necessarily endorsed the content yet.) ** Monday meeting (e-Science) - Geoffrey Fox presentation on 2 core model - People will not converge on one model any time soon. - UK e-Science should be prepared to cover both WSRF & WS-I++ - Still not sure what the objections are to WSRF; WS-I++ ends up trying to do same things in slightly different way. - Proposal to discuss at a later time, perhaps during the F2F, the issue of OGSA dependence on WSRF as well as the opposition to WSRF. (Meanwhile Andrew is following up by talking with some people known to object to WSRF in order to get a better idea of the issues.) - There seem to be both political and technical issues. OGSA-WG should try to understand and work through the technical issues and try to reduce the division, at least within the grid community. - OGSA Profiles: 'conformance' and 'compliance' - Some issues were raised about what is in the BP - Andrew emphasized that the BP is not finished so people who have comments should (and are welcome to) contribute. * Update from WSRF/WSN F2F ** WSRF - Good progress. Many issues resolved. (Issue of which version of WS-Addressing to reference is still open). - A committee draft should be out soon (perhaps within the next couple of months). - There are still some questions with the Service Group definition. ** Base Notification/Eventing There are some places where Base Notification is more complex than WS-Eventing. It is possible to refactor Base Notification but there is no point doing so until the WS-Eventing co-authors respond to the offer sent to them. * F2F logistics - Phone line cost (for bridge) will be covered by Fujitsu. - Meeting room cost will be covered by Andrew. - Network access: - It is too expensive to get a network connection in the meeting room. - There is free wireless in lobby - And high speed network access in the rooms (not free) - Agreed that this is a reasonable compromise. - Projector will be provided by Takuya Mori - Deal is dependent on enough people staying at the hotel. Hiro got a $125/day room rate which is better than the online one. - Hiro has to finalize the deal before people can make room reservations at this rate. And the cutoff date is likely to be very soon. People attending should make sure to reserve soon after Hiro announces it on the list. [In case of questions email Hiro or Andreas.] * Next call: Agreed to cancel Wednesday's call since many people cannot make it due to GlobusWorld. Hiro will send F2F updates to the list.