Dear All,
Please find attached the weekly GridPP Project Management
Board Meeting minutes. The latest minutes can be found each week in:
http://www.gridpp.ac.uk/php/pmb/minutes.php?latest
as well as being listed with other minutes at
http://www.gridpp.ac.uk/php/pmb/minutes.php
Cheers, Tony
________________________________________________________________________
Tony Doyle, GridPP Project Leader Telephone: +44-141-330 5899
Rm 478, Kelvin Building Telefax: +44-141-330 5881
Dept of Physics and Astronomy EMail: [log in to unmask]
University of Glasgow Web: http://ppewww.ph.gla.ac.uk/~doyle
G12 8QQ, UK Video - IP: 194.36.1.32
________________________________________________________________________
GridPP PMB Minutes 174 - 13th June 2005
========================================
Present: Tony Doyle, Jeremy Coles, Robin Middleton, Tony Cass & Steve Lloyd.
Dan Tovey, Peter Clarke & Roger Jones by phone.
Dave Britton & Sarah Pearce by SKYPE
Apologies: Dave Kelsey. (also John Gordon though he reported in at the
start of the meeting)
In attendance for first agenda item - Stephen Burke
1. Deployment & Middleware Issues
=================================
SB was invited to attend this meeting to give his thoughts on issues
relating to Deployment and Middleware.
Deployment UK - SB said there was not enough manpower at some of the sites
and cover was needed at all times. It was problematic to expect sites to
cover for each other. System Managers needed to engage more with the
projects as they do not involve themselves with other areas. TD asked
about induction and JC said new staff are given the standard training but
agreed that that a wider area should be covered and a date should be set
for further induction to take place. TD said PC training resources were
needed for this and asked if this could be arranged at RAL. JC said he
would look into this.(NEW ACTION)
Deployment LCG - SB felt that much of the deployment-level decision making
(below GDB level) was centralised at CERN with limited opportunites for
external input. Limited advanced info. is given and changes to operations
take place on an ad-hoc basis. JC said the Operations Workshop in Bologna
covered a number of these issues but he is waiting on some feedback. TD
said this was a recognised problem: there is multiple country
representation at various boards but there could be more organisation at
lower levels. TD said he would discuss this with TC, JG and Ian Bird to
provide input to construct a recognised sub-structure to try to help the
situation.
LCG vs EGEE - SB said there were differences between the two, which were
growing - LCG were concentrating on getting things done immediately while
EGEE had a much broader remit. He also commented that LCG is more
dominant within GridPP since it is more aligned. TD said that our
priorities were towards LCG as per our remit. He added that if problems
were sorted on the PP Grid, then VOs could be enabled. He also said that
BioMed VO should be supported by all sites.
The PMB AGREED that this should be encouraged at all sites.
Performance - Although 100 sites were green
there were still 45 which were not. SB also felt that Middleware were not
dealing with problems effectively. He did say that VOs were improving
however. TD asked if ordinary users could cope with the system. SB said
there should be more overall effort to support VOs and noted that this was
improving with respect to GGUS and ATLAS following the Bologna meeting.
g-Lite - SB said the expectation was that g-Lite was expected to solve
problems immediately but this was not possible as it was a combination of
independent items and not a single unit. He noted the problems with
CE(Condor-C+CE-monitor), security issues and failure of jobs. There was
poor or no usage of VOMS at present. Data Management was also
problematic, with no replication of data and FTS considered too-LCG
specific. TD said there have been delays with VOMS but people were
working on it now, however the other issues had to be resolved and this
would be discussed at a higher level.
LCG Experiments - SB said that the 4 LHC experiments were all approaching
the Grid in their own way but were each coping. TD said problems were
being recognised and this would be monitored to see if users find the
system more (or less) useful over time.
TD thanked SB for his observations and insights into many of the long-term
problematic issues that are currently faced by the experiments in the
deployment and middleware area.
2. Metrics
===========
JC had sent a draft of the GridPP-PMB-64-Metrics document for reviewing
https://www.gridpp.ac.uk/production_manager/pmb/pmb.html
TD said the document was very good, the only problem being that some
information was being received late. JC explained the structure and said
this was sectioned into the different parts of deployment with a short
analysis followed by GridPP actions. DB said that to include metrics in
the Project Map, he needed to have numbers and a target. JC said there
was no automated process to tell e.g. status of storage elements and it
was not known what experiments will use. TD said as the numbers needed
filtered and processed, the metrics for the project map could be prepared
after the deadline for the OC papers though a sub-set could be reviewed on
Friday. JC said he still had to include references and security incidents
had to be done. TD said caveats would have to be included for many
of the metrics e.g. for the planned inclusion of the EGEE QA RB metrics
for the purposes of conveying info. to the OC.
A spreadsheet would also be included to show where information can be
collected as input to the Project Map.
3. Status of OC input documents
===============================
PMB Executive Summary GridPP-PMB-51-Exec - not yet started
TC LCG Status Report GridPP-PMB-52-LCG - still to be done, to iterate with TD
RM, JG EGEE Report GridPP-PMB-53-EGEE - RM has sent a skeleton to JG for
him to complete his part
DK, JC Deployment Board GridPP-PMB-54-Deployment Board - JC has asked DK
for estimates
RM Middleware/Security/Network GridPP-PMB-55-Middleware/Security/Network -
will be done this week
RJ, DT Applications GridPP-PMB-56-Applications - RJ has sent a version to
DT, TD and DB for comments
TD, JG Tier-1 GridPP-PMB-57-Tier-1/A - a half version has been sent,
waiting for comments from Andrew Sansum
Note: updated significantly during the week.
SL Tier-2 GridPP-PMB-58-Tier-2 - has been sent round for comments
SP Dissemination GridPP-PMB-59-Dissemination -
Note: a draft was circulated on 15th June for PMB comment
DB Project Map GridPP-PMB-60-Project Map - using quarterly inputs, so
other info is needed first. TD said this might be delayed until Tuesday
prior to the OC, but OC would need to be informed. TD and DB agreed to
review next week.
DB Resource Report GridPP-PMB-61-Resources - draft will be done when
some final numbers are received
DT User Board Summary GridPP-PMB-62-User Board - questionnaire being issued
RJ T1-T2 and Analysis GridPP-PMB-63-Analysis - a version has been prepared
JC++ Metrics and "get fit" GridPP-PMB-64-Metrics - draft has been discussed
RM Middleware planning GridPP-PMB-65-Middleware - will be done this week
ACTIONS AT 13TH JUNE 2005
========================
157.2: RJ - review experiment computing models for Tier-1 to Tier-2
interactions in the UK.
ongoing
157.9: RM to report on direction of GridPP in terms of middleware.
ongoing
167.2: JC, with input from Stephen Burke to maintain a list of middleware
issues that should be addressed
ongoing - some iteration has been carried out but further discussion required
167.3: RJ to define UK contacts for each experiment who will respond on
experiment operational issues.
ongoing
167.5 DK to define a possible technical documentation role.
ongoing
170.2: RJ to invite to representatives to talk to the PMB about GANGA status
ongoing - G meeting to be held this week then review
170.3: DT to compile and issue a follow-up questionnaire to UB
ongoing
170.4: TD to speak to COD people about the sound delay in the system
ongoing
172.1: SP to check how much money is available in the budget for
further dissemination possibilities, then to circulate request to UKHEPGRID.
ongoing - BP taking this up. (Grid Cafe proposal was turned down.
New proposal could be submitted to PMB for internal support.)
173.1: JG to resolve allocation of security post for CCLRC
ongoing - the original contender for this post is now moving on to a new
position, the vacancy will be advertised
173.3: SL, RJ & DT to report on production testing of the ATLAS project
done - SL sent Ian Hichliffe's summary:
http://agenda.cern.ch/askArchive.php?base=agenda&categ=a044738&id=a044738s14t9/transparencies
173.4: RJ to include a reference about the status of GANGA in the
Applications Status Report
done - 'though more to be added
NEW ACTION 174.1: JC to look into specialised induction training of
System Managers (to take place at RAL)
Standing Items
==============
SI-1 ALL: to provide any news items and meeting dates for the web page to SP.
BP is liaising with Durham for GridPP13. A news report will be done on
"100 sites".
SI-2 Production Manager's weekly report of issues.
ACTIONS AT 13TH JUNE 2005
========================
157.2: RJ - review experiment computing models for Tier-1 to Tier-2
interactions in the UK.
157.9: RM to report on direction of GridPP in terms of middleware.
167.2: JC, with input from Stephen Burke to maintain a list of middleware
issues that should be addressed
167.3: RJ to define UK contacts for each experiment who will respond on
experiment operational issues.
167.5: DK to define a possible technical documentation role.
170.2: RJ to invite to representatives to talk to the PMB about GANGA status
170.3: DT to compile and issue a follow-up questionnaire to UB
170.4: TD to speak to COD people about the sound delay in the system
172.1: SP to check how much money is available in the budget for
further dissemination possibilities, then to circulate request to UKHEPGRID.
173.1: JG to resolve allocation of security post for CCLRC
174.1: JC to look into specialised induction training of System Managers
(to take place at RAL)
STANDING ITEMS
==============
SI-1 ALL: to provide any news items and meeting dates for the web page to SP.
SI-2 Production Manager's weekly report of issues.
|