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/pmb/minutes/this-week.txt
(as well as being listed with other minutes e.g. 020911.txt).
Cheers, Tony
________________________________________________________________________
Tony Doyle, GridPP Project Leader Telephone: +44-141-330 5899
Rm 482, 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: 194.36.1.234
________________________________________________________________________
GridPP PMB Minutes 47 - 11th September 2002
===========================================
Physical Meeting at RAL
Present: Dave Britton, Tony Cass, Neil Geddes, Tony Doyle, Steve Lloyd,
Robin Middleton, John Gordon.
Apologies: Pete Clarke, Roger Barlow.
ACTION 46.2 : DB to prepare and send agenda to members for this meeting.
DONE
ACTION 46.3 : DB to provide members with updated Project map for this
meeting.
DONE
Project Map
===========
There was a discussion of whether risk assessment should be a separate
document. RM said he would be redoing a risk analysis of EDG for CLRC. The
top levels of this could be used. It was agreed to bear risks in mind
while going through the Map. DB explained that the Project map now has
'sideways' links to supporting documents and web pages. For example 1.1.1
is complete and the link brings up a web page with the evidence. DB would
like a link to something relevant for each completed milestone. The status
date sets the date against which all deliverables are checked and the
boxes are colour coded appropriately according to the key shown below it.
DataGrid
========
Some deliverables are ready but awaiting comments so it is not clear what
due date is appropriate. There was a general discussion of how to set
these dates, especially for EDG software releases. It was decided to
change the due date for release 2 from Sept 30 to Jan 15 reflecting the
changed emphasis from functionality to stability and quality. There are
relatively few milestones left for DataGrid. It was agreed that we are
happy with this. It was discussed whether the date of completion should be
the delivery or acceptance date. SL proposed that it should be the
delivery date with an associated risk that it might not be accepted which
might have resource implications on GridPP. This was agreed. There was a
further discussion of risks. For instance applications require LCG
solutions but are these GridPP risks? It was agreed not to produce a top
level risk document but that each part of the Project Map should have the
associated risks embedded in it.
New Action 47.1 RM to identify top level risks associated with EDG.
CERN LCG
========
DB asked if the milestones are adequate. All boxes are green after
updating with latest dates from TC. It was pointed out that there will be
new milestones. We have overall LCG areas in Project Map. If any more
specific milestones relevant to GridPP emerge we can add them to the Map.
It was agreed that the milestones are OK for now. In a year's time we may
have increased or changed them. It was agreed that in principle previous
documents submitted to the OC should be made public although the PMB
should first assess each document to check that this is OK.
New Action 47.2 TC to produce a Status Reports along lines of the previous
CERN report.
New Action 47.3 SL to make previous OC documents public following review
by PMB.
Applications
============
There will be a meeting on October 9 between DB, PC and NB to review the
quarterly reports. We have a reasonably well developed set of milestones.
The milestones may change now that people are in post and are considering
things better. The Map can be changed in future as long as DB is given a
good reason that can be passed on to the OC. There was a discussion on
whether there should be a document on applications. It was agreed there
should be but it wouldn't necessarily including the Q7 reports. There are
risks due to dependence on LCG (APIs, persistency, convergence with
experiments) and the EDG (stable testbed and functionality, APIs).
New Action 47.4 NB to produce (Q6) summary document.
Infrastructure
==============
For Tier-1/A we need a report including staff deployment issues. JG
explained that the situation much is now better and he has 75 applicants
for the 2 unfilled posts. There will be interviews on 24 September. One of
the matching e-Science posts is in place and one is under offer.
For Tier-2, SL has produced a document. TD asked about Durham and other
smaller groups. It was agreed that SL should contact Cambridge before
sending the document to the CB.
On testbeds, DB asked if Bristol are currently part of it. JG though they
are running the software but not officially in the testbed. There was a
discussion of continued upkeep of EDG releases at institutes and whether
the rollout plan addresses the subsequent releases of EDG versions and
what happens if sites switch to LCG testbeds instead. It was agreed that
this should be a separate item in the Project Map.
New Action 47.5 JG to produce Tier-1/A report including staff deployment.
New Action 47.6 SL to ask Andy Parker about Cambridge before CB.
New Action 47.7 DB to add 4.4.6 saying that by Dec O3 UK will be part of
LCG approved testbed.
Interoperability
================
It was thought that Interoperability and Dissemination should really be
one level lower but TD and DB had agreed to leave it as it was. Various
updates were suggested. TD has produced a list of GridPP Open source
projects and external proprietary software currently in use. It was agreed
that RM should take ownership of the list of produced software and that NB
should look after the list of external proprietary software used.
New Action 47.8 SL to update list of GGF leadership roles (3) and other
items
Dissemination
=============
There was a discussion of papers and how the All Hands abstracts could be
included as conference contributions.
New Action 47.9 SL to add All Hands abstracts and update.
Resources
=========
It was agreed to change 7.1.2 to be number of people not actively in post.
Currently (as of 11 September) there are four posts unfilled (3 RAL, 1
Lancaster). It was agreed that the metric should be 5% of the total
excluding CERN posts. It was agreed to make a separate entry for CERN
posts. In order to value external resources, it was agreed to use the CPU
and disk numbers in SL's Tier-2 document and that JG would then value them
using a an average cost per item i.e. their value would be what it would
cost for GridPP to buy them at RAL now.
New Action 47.10 RM to provide appendix to resource report on travel
spend.
New Action 47.11 DB to obtain CERN resource figures from Chris Eck for
project map.
New Action 47.12 JG to propose nominal value for Tier-2 resources
Feedback from last OC Meeting
=============================
DB had provided an annotated list.
(numbering refers to OC minutes)
3.4. It was not clear what it means. It is not an action item but might
come up at the meeting. The second sentence needs "engaged" added.
3.9. RM produced the current travel spend profile. This is in line with
allocation.
4.2 and 4.3. It was agreed to no longer use the Gantt chart since all
information was accessible via the Project Map.
Action point 6. It was discussed whether the Project Map addresses this.
It was agreed to assume that we are providing what the OC needs.
Executive Summary. Unlike last time the input documents should be in
beforehand this time. It was suggested that TD's achievements list could
form the starting point.
It was concluded that we seem to have addressed all the OC's points.
Documents for OC
================
To be circulated to PMB by 27 September and to OC at least a week
before the 14th October Meeting:
1. Executive summary/latest statement - TD
(to be drafted/circulated on 1st October)
2. Project Map - DB
3. Resource report (DB) + travel (RM)
4. LCG - TC
5. DataGrid Quarterly report - RM
6. Applications quarterly report - NB
7. Tier-1/A report - JG
8. Tier-2 - SL
Other Business
==============
IC Meeting. DB said people should be encouraged to register.
CERN hardware.
TC reported that GridPP had approved the purchase of 300 CPUs and 17TB.
CERN now proposed to fund high performance tape drives instead. TD would
have liked a document showing a need for high performance tape drives. TD
said it is driven by ALICE requirements. DB said it is no problem from a
financial point of view. JG defended tape in the timescale of this
project. SL said tape probably depreciates less than disk or CPU. TC said
that tape is less expensive in manpower at the moment. The proposal was
approved.
ACTION REVIEW
=============
Prior to the meeting, the following actions were performed (as necessary
input).
ACTION 40.3 : DB to finish macros to extract EDG effort and update 2.1.1,
2.2.1, 2.3.1, 2.4.1, 2.5.1, 2.6.1 and 2.7.1 accordingly.
DONE
ACTION 40.6 : RM to email DB confirmation of status of 2.2.5, 2.3.5,
2.4.5, 2.5.5, 2.6.6 and 2.7.4.
DONE
ACTION 40.8 : PC to provide DB with separate ATLAS, LHCb and joint ATLAS/LHCb
milestones.
DONE
ACTION 40.11 : DB to liaise with DC to verify integrated set of SAM/D0/CDF
milestones in 3.4
DONE
ACTION 40.12 : PC to verify 3.5.1 to 3.5.6 and email DB confirmation of status
of 3.5.1 to 3.5.4
DONE
ACTION 40.18 : JG to email DB confirmation status of 4.1.2, 4.2.1, 4.4.2,
4.4.3, 4.5.2 and 4.6.1.
DONE
ACTION 40.20 : TD to verify/confirm 5.1.1 through to 5.4.5.
DONE
ACTION 40.22 : TD to email DB best estimates of current numbers for 5.1.1
through to 5.4.5.
DONE
ACTION 40.27 : DB to provide estimates of current numbers for 7.1.1 through
to 7.3.4.
DONE
ACTION 40.28 : RM to provide input to DB for 7.1.2 and 7.1.4.
DONE
ACTION 40.29 : DB to update status of 7.2. and 7.2.3
DONE
ACTION 40.30 : DB to develop methods for extracting information for 7.3.1
DONE
Following the meeting but prior to the next meeting of the PMB at IC, the
following actions were performed:
ACTION 45.1 : TC to provide TD with some CERN slides for the IC meeting.
DONE
ACTION 47.6 : SL to ask Andy Parker about Cambridge before CB.
DONE
6th Collaboration Meeting. There was an (inconclusive) discussion of
possible dates (subsequently fixed for 30-31st January at Coseners).
ACTION 42.1 : ALL to investigate possible clashes with other projects
during February in order to allow fixing a date for GridPP6.
DONE
ACTIONS AS AT 11TH SEPTEMBER 2002
=================================
ACTION 17.14 : JG revise Tier 1/A staffing plans.
ACTION 35.2 : JG to value additional kit used at RAL and other sites.
ACTION 35.3 : DB to pull additional effort from reporting spreadsheet.
ACTION 35.4 : TD to contact sites which do not yet have a "green dot"
ACTION 37.1 : RM to send email to UKHEPGRID informing them of new codes.
ACTION 37.2 : RM to finalise/distribute Q6 reports to PMB.
ACTION 37.3 : JG to review deputy job description.
ACTION 40.23 : NB to join GGF Applications Group.
ACTION 40.36 : DB check and pull together risk assessments.
ACTION 44.3 : PC to contact EB and TB to re-arrange joint EB/TB meeting.
ACTION 46.1 : SL to send copy of Tier 2 document (after PMB check) to CB.
ACTION 47.1 : RB to identify top level risks associated with EDG.
ACTION 47.2 : TC to produce a Status Report along lines of the previous
CERN report.
ACTION 47.3 : SL to make previous OC documents public following review
by PMB.
ACTION 47.4 : NB to produce (Q6) summary document.
ACTION 47.5 : JG to produce Tier-1/A report including staff deployment.
ACTION 47.7 : DB to add 4.4.6 saying that by Dec O3 UK will be part of
LCG approved testbed.
ACTION 47.8 : SL to update list of GGF leadership roles (3) and other
items.
ACTION 47.9 : SL to add All Hands abstracts and update.
ACTION 47.10 : RM to provide appendix to resource report on travel
spend.
ACTION 47.11 : DB to obtain CERN resource figures from Chris Eck for
project map.
ACTION 47.12 : JG to propose nominal value for Tier-2 resources
|