- pgi-wg - lists.ogf.org

Re: [Pgi-wg] OGF PGI - Review of notes of OGF30 sessions on 26 October 2010 - Counting votes for requirements
by Oxana Smirnova 07 Nov '10

07 Nov '10

Hi, thank you Andre for explaining the procedures, and Morris for being so positively inclined :-) On a sarcastic note, the simplest working code on which a consensus can easily be reached is "return 0". In PGI, we are up to much more challenging goals. Allow me this copy-and-paste from the current PGI charter: <quote> Goals This group will deliver the following document: * Production Grid Infrastructure Roadmap Document (GFD-I) * Provides an overview of missing links between open standards and the fundamental motivation for standardization of the production grid infrastructure profiles * Secure Job and Data Management Profile in Production Grids (GFD-R.P) * Develop a job/data/security profile assuming an deployed computing endpoint is already known by an end-user * Secure Information Profile in Production Grids (GFD-R-P) * Allows a user to discover resources that are appropriate for their request and that they are authorized to access * Secure Accounting Profile in Production Grids (GFD-R-P) * Develop an accounting profile that allows services to securely update a service within information about the resources they have used on behalf of a user. * Secure Monitoring Profile in Production Grids (GFD-R-P) * Develop a profile that allows services to securely record in a service the progress of an activity. </quote> Pretty challenging, IMHO. I only hope "rough consensus" will be possible - and indeed *useful* for code - in each of these cases. Unless of course the "living charter" process will evolve these goals into something easier to achieve. Cheers, Oxana 07.11.2010 20:06, Morris Riedel пишет: > Hi, > > that's what I had in mind too. > >> -- If a group is deadlocked like PGI (or rather if it is running circles >> -- as PGI seems to do), it is the duty of the chairs to push the group >> -- along. In the worst case, if full consensus cannot be reached, a vote >> -- on the available options can lead to rough consensus, which ought to >> -- be enough to get things going again. "Rough consensus - running code" >> -- is the motto for OGF (borrowed from IETF, >> -- http://en.wikipedia.org/wiki/Rough_consensus) > > This particular approach turned already out to be a very effective tool since a couple of months (2 documents out) and with this I see no problem in moving forward and reaching consensus also on the specification level. > > In the initial cycle we did not used this 'tool' trying always to reach a full consensus of all and that was hard. > > > Nevertheless, let's not forget that we produced two documents and increased the mutual understanding. > > > Thanks for this Andre, > Morris > > > >> -- -----Ursprüngliche Nachricht----- >> -- Von: andremerzky(a)gmail.com [mailto:andremerzky@gmail.com] Im Auftrag von Andre Merzky >> -- Gesendet: Sonntag, 7. November 2010 19:01 >> -- An: Oxana Smirnova >> -- Cc: Riedel, Morris; pgi-wg >> -- Betreff: Re: [Pgi-wg] OGF PGI - Review of notes of OGF30 sessions on 26 October 2010 - Counting votes for >> -- requirements >> -- >> -- Hi all, >> -- >> -- On Sun, Nov 7, 2010 at 6:14 PM, Oxana Smirnova<oxana.smirnova(a)hep.lu.se> wrote: >> -- > Hi, >> -- > >> -- > I'd like to point out that my "interesting thoughts" are directly based on >> -- > the PGI group description here: >> -- > >> -- > http://forge.gridforum.org/sf/projects/pgi-wg >> -- > >> -- > This was the mandate of the group when it was approved by the OGF, and it >> -- > explicitly contains the list of relevant standards and specifications, which >> -- > we just re-discovered. It even contains SRM and GridFTP, well in line with >> -- > the stated group's committment to deal with data management - something that >> -- > was contested by the management in Brussels. >> -- > >> -- > Perhaps the group description needs to be updated, if management believes it >> -- > contains controversial statements. What is the procedure for this? >> -- >> -- purely from the OGF procedure perspective, the process would be to >> -- >> -- - draft an agenda update, >> -- - get rough consensus on that update via the mailing list (one week >> -- final call) >> -- - either submit that update to your area director, >> -- - or submit it online to OGF's living charter (which will trigger >> -- the AD as well). >> -- >> -- The update will then be reviewed by the GFSG, and usually accepted if >> -- it is within OGF's mission statement. >> -- >> -- >> -- For PGI, my very humble opinion is that a charter update is not needed >> -- as long as the group is undecided on the explicit way forward -- and >> -- that decision is long overdue. >> -- >> -- If a group is deadlocked like PGI (or rather if it is running circles >> -- as PGI seems to do), it is the duty of the chairs to push the group >> -- along. In the worst case, if full consensus cannot be reached, a vote >> -- on the available options can lead to rough consensus, which ought to >> -- be enough to get things going again. "Rough consensus - running code" >> -- is the motto for OGF (borrowed from IETF, >> -- http://en.wikipedia.org/wiki/Rough_consensus) >> -- >> -- Hope that helps, >> -- >> -- Andre. >> -- >> -- >> -- >> -- >> -- >> -- > Cheers, >> -- > Oxana >> -- > >> -- > >> -- > 07.11.2010 18:01, Morris Riedel пишет: >> -- >> >> -- >> Hi, >> -- >> >> -- >> Interesting thoughts. Indeed. >> -- >> >> -- >> >> -- >> >-- Or will we start all the specifications from scratch? >> -- >> >> -- >> Depends on the rough consensus and majority decisions in the group step by >> -- >> step for each of the specification in question to be >> -- >> profiled/produces by us. >> -- >> >> -- >> >> -- >>> -- Maybe this is also something to clarify on Thursday. >> -- >> >> -- >> Perhaps, but the approach is clear and has been discussed - then with the >> -- >> 'rough consensus' no problem to move forward working on >> -- >> the specifications. >> -- >> >> -- >> >> -- >> Take care, >> -- >> Morris >> -- >> >> -- >> >> -- >> >> -- >> >> -- >>> -- -----Ursprüngliche Nachricht----- >> -- >>> -- Von: pgi-wg-bounces(a)ogf.org [mailto:pgi-wg-bounces@ogf.org] Im Auftrag >> -- >>> von Oxana Smirnova >> -- >>> -- Gesendet: Sonntag, 7. November 2010 17:32 >> -- >>> -- An: pgi-wg(a)ogf.org >> -- >>> -- Betreff: Re: [Pgi-wg] OGF PGI - Review of notes of OGF30 sessions on >> -- >>> 26 October 2010 - Counting votes for >> -- >>> -- requirements >> -- >>> -- >> -- >>> -- Hi Morris, all, >> -- >>> -- >> -- >>> -- I came to think about the process: now that we have the use cases and >> -- >>> have "derived" the requirements (exact set >> -- >>> -- of which can be still argued and prioritised in various manners), is >> -- >>> it time to come back to the specifications? >> -- >>> -- The "strawman" and such? The high-level scheme on the photo is in no >> -- >>> way different from what we had 2 years ago, >> -- >>> -- after all (remember, the group was called "BES/JSDL/GLUE" in 2008), >> -- >>> the circle is complete now. >> -- >>> -- >> -- >>> -- Or will we start all the specifications from scratch? >> -- >>> -- >> -- >>> -- Maybe this is also something to clarify on Thursday. >> -- >>> -- >> -- >>> -- Cheers, >> -- >>> -- Oxana >> -- > >> -- > _______________________________________________ >> -- > Pgi-wg mailing list >> -- > Pgi-wg(a)ogf.org >> -- > http://www.ogf.org/mailman/listinfo/pgi-wg >> -- > >> -- > >> -- >> -- >> -- >> -- -- >> -- Nothing is ever easy...

1 0

Re: [Pgi-wg] OGF PGI - Review of notes of OGF30 sessions on 26 October 2010 - Counting votes for requirements
by Oxana Smirnova 07 Nov '10

07 Nov '10

Hi Morris, all, I came to think about the process: now that we have the use cases and have "derived" the requirements (exact set of which can be still argued and prioritised in various manners), is it time to come back to the specifications? The "strawman" and such? The high-level scheme on the photo is in no way different from what we had 2 years ago, after all (remember, the group was called "BES/JSDL/GLUE" in 2008), the circle is complete now. Or will we start all the specifications from scratch? Maybe this is also something to clarify on Thursday. Cheers, Oxana

2 1

Minutes of OGF30 PGI Session (3) Requirement Prioritization Online
by Morris Riedel 04 Nov '10

04 Nov '10

Hi team, please find the minutes of the final session also online: http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discu… There is a photo in context that has been majorly the outcome of the discussions that can be found here: http://forge.gridforum.org/sf/docman/do/downloadDocument/projects.pgi-wg/do… Comments appreciated during the telcon or via e-mail. Take care, Morris ------------------------------------------------------------ Morris Riedel Deputy Division Leader Distributed Systems and Grid Computing Division Jülich Supercomputing Centre (JSC) Info: http://www.fz-juelich.de/jsc/JSCPeople/riedel

1 0

Minutes OGF30 PGI Session (2) Specification Work Online
by Morris Riedel 04 Nov '10

04 Nov '10

Hi team, please feel free to comment on the following minutes: http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discu… Take care, Morris ------------------------------------------------------------ Morris Riedel Deputy Division Leader Distributed Systems and Grid Computing Division Jülich Supercomputing Centre (JSC) Info: http://www.fz-juelich.de/jsc/JSCPeople/riedel

1 0

Minutes OGF30 PGI Session (1) Presentations Online
by Morris Riedel 04 Nov '10

04 Nov '10

Hi team, please feel free to comment on the following minutes: http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discu… Take care, Morris ------------------------------------------------------------ Morris Riedel Deputy Division Leader Distributed Systems and Grid Computing Division Jülich Supercomputing Centre (JSC) Info: http://www.fz-juelich.de/jsc/JSCPeople/riedel

1 0

Reminder & Agenda: PGI Telcon 2010-11-04; 3:30 pm CET
by Morris Riedel 04 Nov '10

04 Nov '10

Dear PGI team, just a gentle reminder about our usual telcon today essentially debriefing OGF30 and clarifying next steps. (1) Telcon-System and Screen-Sharing http://webconf.vc.dfn.de/ogf-pgi-wg/ No login is required. Just enter as 'guest' with your real world name. (2) Proposed agenda: (a) Discuss EMI PGI involvement (e-mail on the list) (b) Debrief notes and process of OGF30 & clarify next steps! Taking into account also e-mails that have been on the list. (c) AOB Please feel free to add any other topic on the agenda that I may have missed. Your co-chair, Morris ------------------------------------------------------------ Morris Riedel Deputy Division Leader Distributed Systems and Grid Computing Division Jülich Supercomputing Centre (JSC) Info: http://www.fz-juelich.de/jsc/JSCPeople/riedel

1 0

WG: EMI participation in PGI
by Morris Riedel 04 Nov '10

04 Nov '10

All, a "blocked list e-mail" from EMI to the PGI list with statements to consider and to be discussed in the call later today. Take care, Morris Von: Alberto Di Meglio [mailto:Alberto.Di.Meglio@cern.ch] Gesendet: Donnerstag, 4. November 2010 11:09 An: pgi-wg(a)ogf.org Cc: Riedel, Morris Betreff: EMI participation in PGI Dear Morris, dear PGI members, One of the major objectives of the EMI project is to make sure that all EMI services adopt agreed standards and implement them across all supported middleware stacks in a coherent and consistent way. The work done in PGI is for EMI very important and I would like to make sure that EMI can actively contribute to this work and help producing widely agreed open standards. Until now, EMI has not officially taken part in the PGI activities, despite a number of current PGI members are in one way or another related to EMI. The existing representatives from ARC, gLite and UNICORE have traditionally represented the long-term objectives of their respective collaborations. EMI has introduced a different perspective and has requirements and interests that are in some way larger than the sum of its parts. In this context, we feel that asking any one of the existing PGI members to represent EMI would not be correct, both for the members and for EMI itself. I would like to ask you to consider accepting an official EMI representative in PGI to bring the voice of EMI into your activities without ambiguities and with a clear mandate of actively participate to the definition of the PGI specifications on one side and a clear commitment to push the adoption of those specifications in the coming EMI releases. I'm aware that you have issued a call for comments on a list of requirements on which the PGI members are asked to express their opinion. We are probably late for this round, but we would still like the opportunity to contribute to this phase in some practical way. It is of course up to you to decide how this can be done. For the moment, EMI is selecting a person that will be proposed to you as candidate EMI representative should you decide to accept EMI within PGI. I look forward to working with you in the future. Best regards, Alberto ________________________________________ Alberto Di Meglio EMI Project Director CERN - European Organization for Nuclear Research Information Technology Department CH-1211 Geneva 23 Phone: +41 (0) 22 7672389 Fax: +41 (0) 22 7669641 GSM: +41 (0) 76 4872948 E-mail: alberto.di.meglio(a)cern.ch

1 0

meeting notes from Brussels?
by Balazs Konya 03 Nov '10

03 Nov '10

hi Morris, On 2010-10-28 07:11, Morris Riedel wrote: > Hi Oxana, > > minutes will be available soon covering many aspects of your points. > > Take care, > Morris Are the notes already available somewhere? I saw you uploaded the slides. thanks, Balazs -- Balázs Kónya Technical Director European Middleware Initiative www.eu-emi.eu NorduGrid Collaboration www.nordugrid.org Lund University balazs.konya(a)hep.lu.se Institute of Physics, EHEP phone: +46 46 222 8049 BOX 118, S - 221 00 LUND, Sweden fax: +46 46 222 4015

2 1

OGF PGI Session (2) Specification Work Slides Online
by Morris Riedel 02 Nov '10

02 Nov '10

Hi team, I uploaded the two presentations about early specification work in the context of PGI: http://www.ogf.org/gf/event_schedule/index.php?id=2182 I really would like to thank Bernd and Andrew for your efforts and sharing with us your ongoing/planned specification work. Take care, Morris ------------------------------------------------------------ Morris Riedel Deputy Division Leader Distributed Systems and Grid Computing Division Jülich Supercomputing Centre (JSC) Info: http://www.fz-juelich.de/jsc/JSCPeople/riedel

1 0

OGF PGI Session (1) Use Cases and Requirements Slides Online
by Morris Riedel 02 Nov '10

02 Nov '10

Hi team, I uploaded all the slides from our first PGI session at OGF30 last week: http://www.ogf.org/gf/event_schedule/index.php?id=2181 I would like to put your attention to the gLite slides that reached me at 12 at lunchtime and with the session starting already at 13:00 I had no time to check e-mails while my wife wasn't working properly anyway on this day. Nevertheless, Luigi contributed to the requirements before already via the mailing list which was the basis then for the requirement reduction done later. I really appreciate all the efforts of the speakers - very good work! Take care, Morris ------------------------------------------------------------ Morris Riedel Deputy Division Leader Distributed Systems and Grid Computing Division Jülich Supercomputing Centre (JSC) Info: http://www.fz-juelich.de/jsc/JSCPeople/riedel

1 0