This is a static archive of the previous Open Grid Forum GridForge content management system saved from host forge.ogf.org file /sf/go/projects.pgi-wg/discussion.meetings.topc4258 at Sun, 06 Nov 2022 11:28:06 GMT SourceForge : Post

Project Home

Tracker

Documents

Tasks

Source Code

Discussions

File Releases

Wiki

Project Admin
Project: pgi-wg     Discussion > Meetings > PGI TelCon 2009-03-13 - 4.p.m. OGF25 Debrief TelCon > List of Posts
Forum Topic - PGI TelCon 2009-03-13 - 4.p.m. OGF25 Debrief TelCon: (6 Items)
View:  as 
 
 
PGI TelCon 2009-03-13 - 4.p.m. OGF25 Debrief TelCon
(1)
OGF25 Summarizing shortly our sessions, going through the minutes on GridForge

- Morris: Minutes (of every meeting) can be found under discussion section -> meetings
- Morris: Many positive points, but more critics of PGI at OGF25 - however we keep on impress with results rather than 
talk
### TBD Johannes: Precise changes to Davids statement about production Grids in wg session 1 minutes
### TBD All: for changing minutes please use the discussion functionality on gridforge
### TBD Morris: Consensus of the WG has to be achieved before attending the middleware security group meeting (end of 
March)



(2)
Addressing Comments: Liasons with other working groups: SRM, JSDL, BES, etc.
Suggestions: 
SRM -> Alex Sim
JSDL -> Donal Fellows
OGSA-BES -> Andrew GrimShaw/Steven Newhouse
Security -> David Groep

- Steven: unclear what this liasons should be for
- Morris: official contact for knowledge transfer to standardization groups (hopefully participation in PGI as well)



(3)
Addressing Comments: (possible) prioritization to focus efforts:
(i)
Clearly security breaks everything - so lets foster security 
(ii)
bes/glue then
(iii)
data issues


Steven: security might be a block to achieve results in other areas (i.e. bes et al.)
Morris: Priorization on security has been discussed and agreed on
Andrew: was an important aspect in OGSA as well where security has been a weak point for interop



(4)
Addressing Comments: Possible slit-up to sub-groups following the different priorizations in subgroups (initially 
disagreement, but now on the table again)

Morris: already discussed and agreed (many times) that we don't want to have sub-groups
Morris: lack of knowledge behind specific standardization areas is exactly what the problem is, standards in isolation
### TBD All: Attend telcons twice a week for those areas you have interest/experience in, not necessarily chaired by PGI
-chairs
### TBD security related experts: dedicated TelCon on Friday (4.p.m. CET) to pursue security
### TBD computation related experts: dedicated TelCon on Wednesday (4.p.m. CET) to purse OGSA-BES/JSDL problems (
voluntary chaired by Andrew)
### TBD Andrew et al: provide a summary of the documents related to computation from GridForge before Wednesday




(5)
Addressing Comments: production Grids and driving forces of PGI (small fish vs. big fish), manpower issues versus 
production Grids and campus Grids

Morris: we have enormously problems with balancing man power and accurate representation of all production Grids
Morris: There is unfortunately a momentum to loose the close PGI collaboration if this inbalance continues



(6)
Other comments by participants of the TelCon

- Steven: We need precise actions to prepare for the telcon in advance
### TBD: Morris send out actions for preparing the telcons 
### TBD: Morris creates four different E-Mail threads related to the different areas to promote discussions on the list
Post Deleted
Re: PGI TelCon 2009-03-13 - 4.p.m. OGF25 Debrief TelCon
I am afraid that my voice was NOT audible, so I have written down my thoughts here.



(1) OGF25 Summarizing shortly our sessions, going through the minutes on GridForge

Through its discussion thread, I have updated the summary of Standard Working Session 2 :  WSDL files need description 
in human-readable format.



(2) Addressing Comments: Liaisons with other working groups: SRM, JSDL, BES, etc.

SECURITY :

-  At OGF25, there was an 'OGSA Authz Working Group' session led by David CHADWICK and Valerio VENTURI, and described at
 http://www.ogf.org/gf/event_schedule/index.php?id=1532
   I am sorry, but I did NOT attend this session.
   The existence of this session indicates that the OGSA-AUTHZ Working Group is perhaps still alive.
   I am wondering if the work that the PGI group is trying to perform about security should perhaps be performed inside 
this OGSA-AUTHZ Working Group ?
   If PGI decides to perform valuable work on security OUTSIDE the OGSA-AUTHZ Working Group, PGI has to clearly write 
down the reasons.

-  In any way, recommendations on security can be implemented in production ONLY after a thorough review and a written 
approval by JSPG, which is defining the Security Policy of at least EGEE, NorduGrid and OSG.

-  So, beyond David GROEP, I can suggest following contacts :
   -  JSPG :  David KELSEY,  Christoph WITZIG  (I think that both attended PGI sessions at OGF25)
   -  OGSA-AUTHZ :  David CHADWICK, Valerio VENTURI

-  I will attend the MWSG Meeting in Zürich on 30 March 2009 with Morris.



(3) Addressing Comments: (possible) prioritization to focus efforts:

I fully agree.



(4) Addressing Comments: Possible slit-up to sub-groups following the different priorizations

I agree.  I will try to attend both telephone conferences.



(5) Addressing Comments: production Grids and driving forces of PGI

-  As a sustainable infrastructure, EDGeS is a small fish (only a 2-years project with less than 100 people),

-  But a Computing Resources, EDGeS is a BIG fish (potentially MILLIONS of destop computers).

As I already stated, EDGeS has NO brand new Security, Job or Data model, but works on bridges between computing grids.  
Inside its bridges, EDGeS would NOT like to implement a lot of grid-specific plugins (interoperation), but only a few 
OGF-compliant plugins (interoperability).



(6) Other comments by participants of the TelCon

Using telephone number +9900827049931906, my voice was NOT audible.

I order to avoid that in the next conferences, I make following proposals :

-  Use a real SKYPE meeting, and invite telephone number +9900827049931906 for those who can not join SKYPE.
   Below are some Skype pseudos that I already found :
   -  Andrew   GRIMSHAW   uvagrimshaw
   -  Balazs   KONYA      balazs.konya
   -  Moreno   MARZOLLA   moreno7663
   -  Morris   RIEDEL     MorrisRiedel
   -  Gabor    ROCZEI     roczey
   -  Oxana    SMIRNOVA   oxana.smirnova
   -  Etienne  URBAH      etienne.urbah
   -  Johannes WATZL      bratschist
   -  David    WALLOM     d_wallom 

-  Use the video conferencing infrastructure of my IN2P3-LAL institute :
   -  IN2P3 has a Video Conferencing Server (MCU) at ccmcu40.in2p3.fr
   -  LAL has several rooms with H323 equipment, and I can easily book one for each Friday at 16h00 CET.
Re: PGI TelCon 2009-03-13 - 4.p.m. OGF25 Debrief TelCon
Using ordinary skype conference seems to be impossible due to number of participants limitation (AFAIR 10). 
Another possible candidate for meeting is EVO  - http://evo.caltech.edu/evoGate/, http://evo.cern.ch/evoGate/index.jsp. 
Last one seems to be down at the moment or is accessible from CERN only.

Aleksandr Konstantinov
aleksandr.konstantinov @ skype
Re: PGI TelCon 2009-03-13 - 4.p.m. OGF25 Debrief TelCon
Participants of the Telcon:

Andrew Grimshaw
Duane Merrill
Steven Newhouse
Etienne Urbah
Benjamin Henne
David Wallom
Morris Riedel (chair)
Others...

Sorry for missing some folks in my list, please add yourself here.

Take care,
Morris


Re: PGI TelCon 2009-03-13 - 4.p.m. OGF25 Debrief TelCon
Sorry for having missed the meeting, had to tend for some students.

A couple of comments:

- No splitting,please. If the group grows too big to be maintained, it will need to be reduced to its original efficient
 core.

- I do see these Wednesday meetings as a very unfortunate sign of a split that goes against the original group intention
. Let me remind again: the group aims at transferring "academic" versions of BES. JSDL and GLUE into one-two profiles 
that are used in production. Separating BES and JSDL from GLUE and security is counterproductive, IMHO - or at least is 
against the original group intention. I have nothing against Andrew taking care of BES and JSDL, but this is not the 
remit of PGI, where "P" stands for "Production".

 
 


The Open Grid Forum Contact Webmaster | Report a problem | GridForge Help
This is a static archive of the previous Open Grid Forum GridForge content management system saved from host forge.ogf.org file /sf/go/projects.pgi-wg/discussion.meetings.topc4258 at Sun, 06 Nov 2022 11:28:06 GMT