GridPP PMB Meeting 647

GridPP PMB Meeting 647 (23.10.17)
=================================
Present: Dave Britton (Chair), Pete Clarke, Jeremy Coles, David Colling, Pete Gronbech, Roger Jones, Steve Lloyd, Andrew Sansum, Louisa Campbell (Minutes).

Apologies: Tony Cass, Tony Doyle, Dave Kelsey, Andrew McNab, Gareth Smith.

1. Data Management Plan
=======================
This arises re guidelines circulated for the next Consolidated Grant which mentions Data Management Plans. This prompts a review of our plan to determine if it is still fit for purpose and could still be circulated to CG grant-holders. There was a question as to whether the head of grant panel should be consulted first – PG will do so. PC will request any new links to be sent to him for inclusion.
ACTION 647.1: PC will update Data Management Plan.

ACTION 647.2: DB will contact CG-holders and circulate link for Data Management Plan once it has been agreed.

2. OSC Meeting
==============
Date for the next OSC meeting is set for 07 February 2018 in London. This was the latest possible date from those circulated. Documents must be submitted 24 January, which makes the timescale very tight – PG suggests documents must be received before Christmas 2017. The financial plan and other elements fall to PG and the main document will need to be signed off before Christmas so that hard deadlines are in place. Several PMB members will be writing for the Experiment CG (deadline 30 January), CG submission deadline is 13 February.
The first PMB in December would be an appropriate time to revisit this and all actions issued by 11 December.
3. AOCB
=======
a) UKT0 – PC reported Anna, Paul Alexander and Paul Hopkins (from LIGO) were there, amongst others, and defended the proposal for £1.5M to the PPRP. The open session went very well and no challenging questions were received, only some clarification requested. They concentrated on the request for software infrastructure for facilities and Open Stack development, they were reminded that this was only a 4-month Capital call. The panel was unusually large (c. 12 members) and consisted of Jonathan Flynn and others, some of whom do not have a computing background. PC will prompt Tony next week if no decision is received since the start data is proposed for the beginning of December. There were no other bids against the funds.
b) Tomorrow DB and PC are attending a meeting by BEIS for heavy computer users within the research councils to discuss commercial cloud usage. This is a large meeting (c. 30 attendees) and scheduled talks that will not discuss costs. PC and Jeremy Yates will be giving talks. PC noted that BEIS require to tick boxes for inward investment, but spending on H/W is cost-neutral. DB and PC will report to the PMB next week
c) DB and PC will attend Geneva next week and present on the UK position on computing strategy – this follows the overview board on Friday this week. They will update the PMB next week.
d) PG noted that the deadline for quarterly reports has now passed and he will email appropriate people shortly to request these.

4. Standing Items
===================

SI-0 Bi-Weekly Report from Technical Group (DC)
———————————————–
The next meeting is Friday, nothing to report.

SI-1 Dissemination Report (SL)
——————————
The post at Imperial is going ahead and Janusz will start very soon – DC will check and circulate an email to confirm. The agreed procedure is that Imperial will pay Janus and invoice QMU. The LZ report that was circulated requested GridPP commitment, one requirement is the implicit use of Dirac and that this will be supported, so we need to ensure manpower availability to support. Currently Simon and Daniella are predominantly working on DIRAC support – Daniella’s service tasks to the GridPP is to run the production of Dirac Server at Imperial. DB recognised this is a lot of work and GridPP5 does not support this, so a plan needs to be clarified before a commitment can be made.

SI-2 ATLAS Weekly Review and Plans (RJ)
—————————————
Nothing significant to report.

SI-3 CMS Weekly Review and Plans (DC)
————————————-
DC noted CPU efficiency task force is finding interesting things and he will report on this when appropriate.

SI-4 LHCb Weekly Review and Plans (PC)
————————————–
AM not present, no report submitted.

SI-5 Production Manager’s report (JC)
————————————-
Here are some updates/news items from ops.

1. The T2 figures for September are as follows:
• ALICE
o Oxford: N/A: N/A
• ATLAS
o Liverpool: 79%:89%
o Sheffield: 88%:88%
• CMS
o All okay
• LHCb
o Liverpool: 80%: 100%
Issues encountered:
o Sheffield: One WN had a CVMFS access problem.
o Oxford: BDII is dual stack. New switch/router changed IPv6 setup. CERN’s BDII could not see site. Site lost from REBUS.
o Liverpool moved to a new machine room in mid-September. Also the university firewall started to reject SSL connections which impacted ATLAS and LHCb reliability.
2. HEPiX took place last week: https://indico.cern.ch/event/637013/timetable/#all.detailed. The meeting covered regular/expected topics with many site reports.
3. There was a GDB on 11th October: https://indico.cern.ch/event/578991/. This included an update on the “System performance and cost model working group”. Note that Gareth Roy and Andrew Sansum are members. A draft mandate was shared.
4. The UK NGI security team are following up on a number of pakiti flagged issues at several sites. (Mentioned to ensure the effort gets some visibility).
5. ROD work has generally been quiet with just a few transient issues.
6. There will be a WLCG HEP Software Foundation workshop in Naples 26th-29th March 2018. The focus will be activity areas identified in the Community White Paper.
7. NA62 are running a large campaign which will complete by the end of December. Additional sites have been asked to support the VO and many have responded – some are concerned that the VO does not currently run on CentOS7
DB noted on the cost model working group he is happy AS, Gareth Roy and Alessandra (and Graeme Stewart) are on the group and they will raise issues about the direction taken. He noted some aspects are applicable in one country but not in another, e.g. funding models differ between each country and it is important the UK is represented to take account of this.
AS noted a new staff member starting today (Darren Moore) who will replace Gareth’s operational role – Gareth will gradually fade out of his production manager role. Tier1 post has been approved and will soon be advertised. Andrew (Primary team leader) resigned with very short notice so in the short term we are reduced to one DB admin and Gordon Brown (previously DB team leader) has returned and leading facilities programme – he will support this in the short term. DC noted that there is an interview date set for the CMS role.

SI-6 Tier-1 Manager’s Report (GS)
———————————
Castor:
There were problems with CMS Disk becoming full at the end of last week. The draining of older disk servers (2012 generation) – partly in anticipation of CMS moving to Echo had been taking place and this had taken CMS to just below pledge. This was causing significant problems for CMS and on Friday five of this generation disk server were put back into service (a total of a bit over 400TB). (Although one disk server would have taken us well above pledge – just adding one would have resulting in hotspotting. Hence five were added.)

Echo:
Still cautiously re-distributing data onto the 2015 capacity hardware.

Networking:
On Wednesday morning one of the three links that makes up the OPN connection was re-routed. (Circuit LHC-1). A cabling change was made by Janet staff at Harbour Exchange at the same time as one made at RAL.

SI-7 LCG Management Board Report of Issues (DB)
———————————————–
Previous meeting was cancelled and there is no MB report.

SI-8 External Contexts (PC)
———————————
Nothing to report.

REVIEW OF ACTIONS
=================
638.2: AS will check when equipment is due to become obsolete and investigate legal and manpower of donation to the African Data Centre for Bioinformatics and Medical Research. (Update: AS is looking into how this may impact Global Challenge Research Fund – GCRF – which would involve a cross-Council bid). Ongoing.
644.2: PG and AS will document plans and costings for the remainder of GridPP5 taking account of the Oracle tape issues experienced. Ongoing.
644.3: AS put together a starting plan for staff ramp-down. Ongoing.
644.4: AS will progress capture of funds for Dirac with Mark Wilkinson. Ongoing.
644.6: (during GridPP39 meeting) DB will ascertain what Biomed do and how they use the Grid. (Update: Holloway then Manchester and Imperial are biggest representatives – DC and AM will investigate this and ascertain whether our input is being explicitly credited).

ACTIONS AS OF 23.10.17
======================

638.2: AS will check when equipment is due to become obsolete and investigate legal and manpower of donation to the African Data Centre for Bioinformatics and Medical Research. (Update: AS is looking into how this may impact Global Challenge Research Fund – GCRF – which would involve a cross-Council bid). Ongoing.
644.2: PG and AS will document plans and costings for the remainder of GridPP5 taking account of the Oracle tape issues experienced. Ongoing.
644.3: AS put together a starting plan for staff ramp-down. Ongoing.
644.4: AS will progress capture of funds for Dirac with Mark Wilkinson. Ongoing.
644.6: (during GridPP39 meeting) DB will ascertain what Biomed do and how they use the Grid. (Update: Holloway then Manchester and Imperial are biggest representatives – DC and AM will investigate this and ascertain whether our input is being explicitly credited).
647.1: PG will update Data Management Plan

647.2: DB will circulate link for Data Management Plan once agreed.