Etienne, all, GridForge is constantly down for me and/or too slow, therefore I distribute the minutes via the list. Just a little reminder about our tasks to check with the existing specifications we discussed during the telcon. Participants Emmanouil, etienne, Luigi, Michaela, Oxana, Steve, Mark (a) Debrief Minutes from last meeting -> OGF30 Session 1: Requirements and Use cases http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discus... Session 2: Early adopters specification work http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discus... Session 3: Summary Session http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discus... (b) Discuss EMI PGI involvement (e-mail on the list) 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. Im 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 Oxana (ECB) : Common representative towards PGI (who that be is in question), key issue whether PGI sees this as useful, because of the discussions in Brussels to have some one in the room to express the EMI views on PGI. We are committed to do a specification of PGI that's why the project EMI should also have a voice. TBD: Postpone for one week the final correction of use cases requirements? TBD : no strong objections - postpone the final correction for one week - statement from one EMI member about the requirements. Bottlom line : It is expected that the major route will not change... (c) Debrief notes and process of OGF30 & clarify next steps! Starting from the photo to where we going now.... Taking into account also e-mails that have been on the list... Oxana: High-level picture is correct - transfer photo in a chart and try to map it and take everything into account - roughly mapping from requiements Morris: discussed dependency on the GLUE2 XML Rendering --> Focus first on this, because not ready yet and we have to push this Morris: Roughly go forward with the GLUE2 specification (rough information model) - rendering needed for normative specification Morris: GLUE2 XML Rendering needed for JSDL XML-based schema, BES endpoint that will be described XML renderings Morris: GLUE2 session at OGF30 - they work on it and it will finished even this year (not public comments... not sure) Etienne : The simple existence of GLUE2 entities and attributes is enough to begin real work. Renderings are NOT so normative: there will be 3 of them ! Circular dependency problem Emmanouil --> SQL Rendering Goal: Take the Excel requirements we agreed on and then put them into the categories e.g. text...mentioning GLUE2,... --> GLUE2 and Specification Areas - and come up with first ideas (roughly) of how to put in specification, taking existing specifications into account and mark gaps Specification Areas: GLUE2 (1) - TBD: Requirements in this area agreed on - Who : Etienne JSDL 2.0 (2) - TBD: Requirements in this area agreed - Who : Mark (is a chair of JSDL group) - Parameter sweep, JSDL 1.0,... JSDL SPMD 2.0 (3) - TBD: Requirement - Who : Morris (expertise in HPC systems) - Etienne: take specification of HPC-Basic Profile --> HPCBPA specification into account, check related work For SPMD, this is GFD.115 PGI Applications Profile, or similiar (4) - TBD: Requirement - TBD : Steve Etienne URBAH (LAL, France): For Applications, this is GFD.111 BES 1.1/2.0 (5) - TBD: not overcommit and postpone a week - TBD : Luigi Tasks: (1) Map the requirements to the corresponding section GLUE2, JSDL2, JSDL SPMD (2) Chart with the understanding of these next steps: in the next call check interdependencies Security : Delegation No time to discuss e-mails on the list --> next week with having a better understanding also of the outcome of OGF30 (3) AOB TParticipants Emmanouil, etienne, Luigi, Michaela, Oxana, Steve, Mark (a) Debrief Minutes from last meeting -> OGF30 Session 1: Requirements and Use cases http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discus... Session 2: Early adopters specification work http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discus... Session 3: Summary Session http://forge.gridforum.org/sf/discussion/do/listPosts/projects.pgi-wg/discus... (b) Discuss EMI PGI involvement (e-mail on the list) 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. Im 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 Oxana (ECB) : Common representative towards PGI (who that be is in question), key issue whether PGI sees this as useful, because of the discussions in Brussels to have some one in the room to express the EMI views on PGI. We are committed to do a specification of PGI that's why the project EMI should also have a voice. TBD: Postpone for one week the final correction of use cases requirements? TBD : no strong objections - postpone the final correction for one week - statement from one EMI member about the requirements. Bottlom line : It is expected that the major route will not change... (c) Debrief notes and process of OGF30 & clarify next steps! Starting from the photo to where we going now.... Taking into account also e-mails that have been on the list... Oxana: High-level picture is correct - transfer photo in a chart and try to map it and take everything into account - roughly mapping from requiements Morris: discussed dependency on the GLUE2 XML Rendering --> Focus first on this, because not ready yet and we have to push this Morris: Roughly go forward with the GLUE2 specification (rough information model) - rendering needed for normative specification Morris: GLUE2 XML Rendering needed for JSDL XML-based schema, BES endpoint that will be described XML renderings Morris: GLUE2 session at OGF30 - they work on it and it will finished even this year (not public comments... not sure) Etienne : The simple existence of GLUE2 entities and attributes is enough to begin real work. Renderings are NOT so normative: there will be 3 of them ! Circular dependency problem Emmanouil --> SQL Rendering Goal: Take the Excel requirements we agreed on and then put them into the categories e.g. text...mentioning GLUE2,... --> GLUE2 and Specification Areas - and come up with first ideas (roughly) of how to put in specification, taking existing specifications into account and mark gaps Specification Areas: GLUE2 (1) - TBD: Requirements in this area agreed on - Who : Etienne JSDL 2.0 (2) - TBD: Requirements in this area agreed - Who : Mark (is a chair of JSDL group) - Parameter sweep, JSDL 1.0,... JSDL SPMD 2.0 (3) - TBD: Requirement - Who : Morris (expertise in HPC systems) - Etienne: take specification of HPC-Basic Profile --> HPCBPA specification into account, check related work For SPMD, this is GFD.115 PGI Applications Profile, or similiar (4) - TBD: Requirement - TBD : Steve Etienne URBAH (LAL, France): For Applications, this is GFD.111 BES 1.1/2.0 (5) - TBD: not overcommit and postpone a week - TBD : Luigi Tasks: (1) Map the requirements to the corresponding section GLUE2, JSDL2, JSDL SPMD (2) Chart with the understanding of these next steps: in the next call check interdependencies Security : Delegation No time to discuss e-mails on the list --> next week with having a better understanding also of the outcome of OGF30 (3) AOB Your co-chair, Morris
-- -----Ursprüngliche Nachricht----- -- Von: Etienne URBAH [mailto:urbah@lal.in2p3.fr] -- Gesendet: Freitag, 5. November 2010 20:39 -- An: Riedel, Morris; Johannes WATZL -- Cc: pgi-wg@ogf.org -- Betreff: OGF PGI : Notes of telephone conference held on 04 November 2010 -- -- Morris and Johannes, -- -- Thank you in advance for publishing quickly the notes of the OGF PGI -- telephone conference held on 04 November 2010. -- -- Best regards. -- -- ----------------------------------------------------- -- Etienne URBAH LAL, Univ Paris-Sud, IN2P3/CNRS -- Bat 200 91898 ORSAY France -- Tel: +33 1 64 46 84 87 Skype: etienne.urbah -- Mob: +33 6 22 30 53 27 mailto:urbah@lal.in2p3.fr -- -----------------------------------------------------