=================================
Present: Dave Britton (Chair), Pete Gronbech, Andrew Sansum, Jeremy Coles, Steve Lloyd, Gareth Smith, Peter Clarke, Dave Colling, Andrew McNab, Roger Jones, Pete Clarke, Claire Devereux. (Minutes – Louisa Campbell)
Apologies: Tony Doyle, Tony Cass, Dave Kelsey
1. ACAT Travel (DB)
====================
DB suggested there should be no real concerns with approving ACAT travel as there are likely to be only two requests: one from RAL to support somebody from SCD who is not technically GridPP but is talking about energy efficiency at Tier-1 so this is directly relevant. The other relates to a talk from Glasgow which DB will probably present since he will be in attendance as a member of the International Advisory Committee. It is not yet clear whether others from the UK will be attending – because the LHC experiments only name speakers after the abstracts have been accepted. As the travel will be undertaken in January and therefore within this year’s budget it was agreed that 50% cover of travel from GridPP was appropriate.
2. Hosting CHEP (DB)
==============================
Following on from San Francisco in 2017, CHEP in April 2018 should rotate to a European host. DB enquired whether we should consider proposing to host CHEP following repeated requests from Glasgow Conference Bureau offering to assist in drafting a letter of intent (due in January 2016). Hosting had also been considered by Imperial, but not yet explored further. If Glasgow were to host it would save on travel costs, but organisation would be a substantial commitment of more than 1 year of LC’s time, with c.500+ delegates it is essential for the event must be extremely well run and organised. Glasgow Conference Bureau are very proactive and provide Civic reception and some marketing/signage support. We have to balance whether 1 year of work is beneficial. There needs to be block booking of hotels 1 year in advance and then Chairs appointed for sessions. CD was a member of the local organising committee for EGI, in Manchester the conference bureau dealt with block booking rooms and the detail. This amounted to less than 1 year of work – 7 parallel sessions and a large conference dinner as well as social events. No sponsors were secured for this but it required a great deal of modelling relating to fixed and variable costs to set appropriate delegate rates – leading to constant checking of projected models to ensure the event did not operate at a loss or a substantial profit. In other conferences the IAP underwrite this and also undertake a fraction of the organising. Alternatively a local group could do this, but there is a substantial financial risk. It was agreed that hosting in Glasgow was cheaper than London or other potential venues. AS noted no other European bids were submitted last time and considerable pressure was applied to find a host. CD suggests to check with Glasgow Conference Bureau the amount of help they would provide.
DB noted the conference bureau provide information on venue options and trip options, they also provide a complementary Civic Wine Reception. If the conference was to take place on-campus the university conference service (CVSO) arrange venue hire and online registration at a cost of £35 per person as well as all catering. It should be noted VAT at 20% would be applied to all venues and catering charges if the event was to be hosted off-campus. DB confirmed the campus is challenging to navigate due to ongoing heating works and there is substantial infrastructure plans for the next 2 years, including the renovation of the Kelvin Building and demolition and rebuilding of the Western Infirmary site, amongst others. CD offered to share planning spreadsheets with costs etc. CD asked if a committee decides on the final programme, DB confirmed we would need a local organising committee (in reality a UK organising committee) if the proposal to host were to proceed. DB and LC will make some enquiries to determine whether hosting is a viable option. There is a venue in Edinburgh close to the castle that may be able to accommodate the number of delegates. DC mentioned that at Imperial the Royal Geographical Society lecture theatre can accommodate 750 and parallel sessions could be accommodated on campus. PMB agreed there is a great deal of work with no real tangible benefits, though a small profit may be achieved. Should hosting the event be agreed it would either replace or incorporate the GridPP meeting in April.
ACTION 581.1 DB and LC to make enquiries on the viability and potential benefits of hosting CHEP conference at Glasgow in 2018.
ACTION 581.2 CD to provide DB and LC with copies of relevant spreadsheets, budgets and planning information from recent EGI conference.
3. AOCB
=======
a) Amazon item from Tony Hey
————————–
DB stated Tony has not contacted him direct but he suggested to Alistair to have a discussion with Tony and DB can be included in this if required. AS has spoken to Tony as it followed on from Tony’s interest in Hartree, AS had mentioned Particle Physics engaging in a lot of work on Cloud and running on HPC resources etc. DB mentioned the possibility of making use of opportunities regarding Hartree. It is thought most likely Tony is exploring options and engaging in a number of discussions which is encouraging.
b) PG issued a reminder for all PMB members who have not already done so to submit their reports to him.
ACTION 581.3 ALL members who have not already done so should submit reports to PG.
4. Standing Items
===================
SI-0 Bi-Weekly Report from Technical Group (DC)
——————————————-
A technical meeting is scheduled for Friday and DC will provide a report next week.
SI-1 Dissemination Report (SL)
——————————————-
###GridPP Website 2.0
####PMB Minutes functionality
The PMB minutes functionality has now been fully implemented. The list of minutes may be viewed at [1]. The archives are listed in the sidebar on the left. Comments and suggestions welcome. Once finalised, TW will write up brief guidelines for uploading the minutes (which will require the user to log-in with their certificate, all handled my AM’s X509 WordPress plugin).
####Switching to www.gridpp.ac.uk
TW believes there is nothing else to be done before a “soft launch” to www.gridpp.ac.uk – i.e. we make the switch but do not widely publicise for a week or so to iron out any minor bugs (e.g. incorrect URLs, certificate issues). Suggested timeline:
* Tuesday 1st Dec: soft launch
** Switch new site to www.gridpp.ac.uk
** Move old site to archive.gridpp.ac.uk and protect with grid certificate verification; keep editable for now during the transition.
** PMB to check for bugs/issues with new site at www.gridpp.ac.uk
* Thurs 4nd Dec: announce switch on TB-SUPPORT
** Solicit feedback/user issues and fix/advise as appropriate.
* Monday 7th Dec: Announce official launch of new website:
** Update social media
** Update STFC et al. with new links, info for www.stfc.ac.uk
* Friday 18th Dec: make archive.gridpp.ac.uk read only.
What do people think?
[1] https://vm36.tier2.hep.manchester.ac.uk/collaboration/pmb-minutes/
Agreement was reached to announce the new website, possibly tomorrow after the Tuesday Ops meeting.
Tom has requested PMB approval to confirm whether staff are working with GridPP if he is being asked for authorisation by RAs. Agreement was reached that Tom is so authorised.
ACTION 581.4 AMcN to liaise with Tom on announcing the new website.
SI-2 ATLAS Weekly Review and Plans (RJ)
—————————————
Nothing of note to report. There has been a good deal of activity looking at consistency checking on Storage. Most of the ATLAS team are in Barcelona this week for information exchange. There was some discussion on Cloud to provide information to Alistair.
SI-3 CMS Weekly Review and Plans (DC)
——————————————-
Nothing significant to report.
SI-4 LHCb Weekly Review and Plans (PC)
——————————————-
Only UK-specific point is Queen Mary to become Tier-2d for LHCb. This will be in place early next year.
SI-5 Production Manager’s report (JC)
——————————————-
1. We have no open regional tickets. Our open GGUS tickets had decreased to below average at 18 just over a week ago, but have increased again to 33 with various ATLAS requests (mainly disk consistency checks) and a push for enabling pilot jobs for our smaller VOs in the UK.
2. The future use cases for the Information System (mentioned at the PMB last week) are now written up in this document: https://espace.cern.ch/WLCG-document-repository/Technical_Documents/WLCGFutureISUseCases_1.3.pdf. The GridPP usage is mentioned in section 6.2 together with our desire to use REBUS and remain with a federated Tier-2 view.
3. One of the most active discussions in the UK community this last week has been about use of certain tools for configuration managamenet (in particular Foreman and Puppet).
4. EGI is currently reviewing WMS provision and usage. This is very relevant for us as we still run a number of instances but on old and possibly unsupported hardware.
SI-6 Tier-1 Manager’s Report (GS)
——————————————-
General:
– Generally a quiet week last week.
Castor:
– No significant changes in last week. The parameter change reported last week (increase number of nameserver database connections) for LHCb has improved, but not fixed, the low level problem whereby some batch jobs fail to write their results into Castor.
– There were failures of three disk servers (all in AtlasDataDisk) over the weekend. All now back in service – temporarily in read-only mode.
– Test of Castor 2.1.15 underway. The plan is to install this in the New Year. (Castor 2.1.15 is already running on the tape servers, but these have no connection to the databases).
Networking:
– We continue keeping a close watch on some low-level packet loss within our network.
– We also continue with the changes needed to remove the old ‘core’ switch from the network. We plan to move the link to the UKLight router on Tuesday of next week, and the link to the Castor headnodes the day after (there will be a ‘warning’ for an hour on Castor for this).
– Last week I reported an increase in external traffic round the bypass link which had been saturating. However, usage levels dropped back and have stayed lower since (around 1 – 2 Gbit/sec).
– As reported last week the OPN has been busy in the last couple of months.
Batch:
We are implementing a changed algorithm for the draining of worker nodes to make space for multi-core jobs. The new version allow “pre-emptable” jobs to run in the job slots until they are needed. An example of these pre-emptable jobs is the Atlas Event Service.
In this regard there was a successful test of the Atlas Event Service using the S3 interface to ECHO a couple of weeks ago.
Procurement:
No further update regarding the procurements. As reported last week the tenders are set to close on 18th December. It is possible there will be a request for an extension of the disk tender by a week. This was the later one to be issued.
SI-7 LCG Management Board Report of Issues (DB)
———————————————–
DB noted the next meeting has been moved to 8 December. DB is travelling from the DIRAC meeting and PG will be at Heathrow at 3pm awaiting a flight departure to Canada at 5pm so both will be able to attend the meeting for at least one hour.
REVIEW OF ACTIONS
=================
578.2 AM and JC to investigate EGI community platforms to see whether VAC and possibly DIRAC could be registered. Ongoing.
578.8 DB will consider reputation risks of inadequate support of new VOs. The issue is addressed through monitoring interaction between new groups and existing infrastructure at UKTO Meetings. Support and monitoring is also facilitated through meetings that all new VOs must attend and Tuesday Ops Meetings. Done.
579.4 DB will adjust the wording on visit notices to make clear that staff are required to make the most cost-effective travel and accommodation arrangements. Done.
580.1 AS will look at tape planning to determine if we want to take forward increasing ALICE space and if this can be easily accommodated in all the scenarios. (To be discussed at the Resource Meeting on 16.12.15) Ongoing.
580.2 AMcN will send round an email to PMB to ask for suggestions for inclusion in a document for WLCG to produce relating to REBUS. Done.
580.3 AS will investigate discrepancy between what Tom is being sent and realities for certificates/authentication. Done.
580.4 After DB updates Visit Notices wording and send PMB message to use site immediately to test any issues and if all works well fully move ahead with new website. Done.
580.5 JC to provide information on how much resource Pravda was found to be useful. (See Appendix I) Done.
ACTIONS AS OF 30.11.15
======================
578.2 AM and JC to investigate EGI community platforms to see whether VAC and possibly DIRAC could be registered. Ongoing.
580.1 AS will look at tape planning to determine if we want to take forward increasing ALICE space and if this can be easily accommodated in all the scenarios. (To be discussed at the Resource Meeting on 16.12.15) Ongoing.
581.1 DB and LC to make enquiries on the viability and potential benefits of hosting CHEP conference at Glasgow in 2018.
581.2 CD to provide DB and LC with copies of relevant spreadsheets, budgets and planning information from recent EGI conference.
581.3 ALL to members who have not already done so should submit reports to PG.
581.4 AMcN to liaise with Tom on announcing the new website.
Appendix I
Response to Action: “580.5 JC to provide information on how much resource Pravda was found to be useful.
Although I have not heard specific numbers back from the primary Pravda user (Tony Price), we have been able to extract some information from DIRAC as shown below. It is a cumulative plot so the answer is that they used about 6000 HS06 days of CPU time.
Jeremy”