=================================
Present: David Britton (Chair), Tony Doyle, Roger Jones, Pete Gronbech, Dave Kelsey, Tony Cass, Gareth Smith (Minutes – Louisa Campbell)
Apologies: Steve Lloyd, Andrew Sansum, Jeremy Coles, Andrew McNab, Dave Colling, Claire Devereux, Pete Clarke
1. GRIDPP5 – next steps
=======================
We need to respond to the STFC on requirement to do 10-day turnaround of GRIDPP5 grants:
Need to resubmit costs for institutes and RAL for normalised 90% of flat cash, but not through the JES system. DB has a template that was sent to PG who has used this previously. DB stated that the STFC council meeting is at the end of September and the grant could be received at the beginning of October. DB noted that there was no time limit stated to him and TD suggested it should be 10-day turnaround, the consolidated grants information has already been submitted.
PG noted that most sites submitted JES forms which won’t change in the 90% scenario, but he will contact the sites where this is not the case. DB stated his partial buy-out time would be reduced and JES figures should change. PG will cost out the figures based on what he already has and then pass onto the sites who should advise if they disagree with his figures – they must inform him by Friday 28/8/15.
DB wrote to RJ as ATLAS representative to confirm whether they want 0.5 FTE in 2018/9 moved from Sheffield to Manchester.
DB noted he would be very busy in various meetings over the next few days.
ACTION 572.1
PG to liaise with institutes and check costs and corresponding RAL figures then do calculations based on staff costings – these must be done by Friday 28.8.15 or early w/c 31.8.15 at the latest. Also to check if ATLAS propose any changes before finalising.
b) Consumables
Part of normalisation – STFC want to know what the spend on consumables is. Tier-1 spend does not have consumables as such. There is a very very small sum called consumables (£5k/yr) for the project leader.
ACTION 572.2
DB to send note about consumables to STFC.
c) Normalise travel budget
GRIDPP has always strictly controlled travel spend, however, it is no great surprise this would be raised. DB noted this currently travel works out at £3.9K per FTE per annum. DK noted that our travel requirements don’t scale with the number of people; it scales with areas of responsibilities and we are trying to preserve those in the reduced funding proposed. Often members have to travel to other countries which can be expensive (more so than experiments, who’s travel is mostly just to CERN), discussion on details of costs of necessities of travel. It was noted that phone meetings are conducted wherever possible and on some occasions collaborative meetings are essential and unavoidable. It was highlighted EGI and WLCG, £35,000 international obligation that does not scale with FTE’s, eg experiment collaboration meetings cost c. £14,000 in addition to WLCG meetings. DB asked DK to resend his list so that a breakdown of costs could be prepared and categorised/classified as ‘core business’ in percentage terms to make a case without being bogged down with too much detail.
ACTION 572.3
DB will make a case noting all of the above and try to pare back travel costs as much as possible but taking account of the above issues. He will liaise with DK before sign-off with PMB.
d) Case for FEC on 12 posts
Follow up on an email from Sarah Verth, we have been requested to justify FEC on 12 specified posts. Since we provided no information on individual posts in the proposal or to the review committee, it is not clear how these specific posts were selected for scrutiny. In particular, it is odd that 3 or the 12 posts are at Glasgow. We need provide ¼ to ½ page per post. DB will draft the outline for these posts in generic document then bullet-point post-specific information for each. Eg, XX has a PhD, publication track record, research profile and team leader experience; YY has a PhD, national research profile and leadership experience; ZZ is Atlas author, etc. Noted that not all post holders have PhD but job descriptions typically have wording that states “PhD or equivalent experience”. This will form a template that the other institutions can add content about their posts. PC noted that GRIDPP staff are not technicians as they are involved in international issues, self directed and function as independent researchers who are required to engage in high level problem-solving.
ACTION 572.4
DB will construct a generic paragraph for the introduction for 3 cases based on the Glasgow posts with a set of headings and circulate to others.
2. GRIDPP35 Meeting
===================
DB reports the agenda is in good shape, PG worked on this. Hotels are booked, collaboration dinner is reserved with numbers up to over 70 and menus were discussed. Clarification on what hotels delegates are booked into is required and DB will send out a list to confirm (first 30 to reserve places were booked into the Holiday in Lime St and others into Holiday Inn at Albert Dock). The list is available on the website so an announcement should be put out to advise. The costs were £82 per night for group booking, others who booked individually received £62 per night rate. Delegates are required to pay for their own expenses.
ACTION 572.5
DB will email UK GRIDPP about the costs for the hotels and confirm allocated hotels.
3. AOCB
=======
a) RCUK Cloud Working Group (PG) was discussed. DB mentioned he has not yet had the opportunity to read JC’s circulated paper. DC is on a group chaired by Jasmine and DB has suggested Ian Collier be considered. Unclear whether DC has had the opportunity to propose this as he is on annual leave.
b) PPAP Community Meeting – two talks will be given, one on 24th Sept by DB and the other on 25th by PC. They will liaise to determine the appropriate content in each.
c) Janet e2e Pert Workshop – Tim Chown coming to GRIDPP35 to give a talk as a precursor to the workshop; we will also probably have talks on IPP6 and Pefsonar. Good relationship with JANET on networking. Tim is doing a very good job on passing information and we are well integrated.
d) SLs metric page – how do we measure work done using VAC? This matter is not urgent right now in light of the pressing matter of STFC reporting, but should be discussed at the next face-to-face PMB meeting in Liverpool on 9 September.
4. Standing Items
==================
SI-0 Report from Development (Cloud) Group
——————————————-
DC absent, no report
SI-1 Dissemination Report
——————————————-
SL absent, no report
SI-2 ATLAS Weekly Review and Plans
——————————————-
JR noted nothing major to report
SI-3 CMS Weekly Review and Plans
——————————————-
DC absent, no report
SI-4 LHCb Weekly Review and Plans
——————————————-
PC and AMcN absent, no report
SI-5 Production Manager’s report
——————————————-
JC absent, no report
SI-6 Tier-1 Manager’s Report
——————————————-
GS reported as follows:
Operations:
Networking:
1) Tier-1 Routers: These have worked well since the intervention with the vendor on the 4th August. As part of the earlier tests we had been using a borrowed router as the primary – and on Tuesday, 18th Aug, we successfully (and without network interruption) swapped our own unit back in.
2) Following the achieving of that network stability we are looking at some further network changes that need to be done. Over the next month we will make some connection changes at a lower level to enable the removal of our old ‘core’ switch.
3) As part of an investigation into a low level packet loss we switched to use different fibres to connect from the Tier-1 core network to the UK Light router – although this does not look to have improved the loss rate.
Castor:
4) We are just starting on the migration of our Castor disk servers to SL6. After discussion with VOs we will do this via a rolling update – with Castor services announced “At Risk”. We have also scheduled further outages to Castor for the update of the Oracle database to version 11.2.0.4.
5) A change to the choice of Linux kernel I/O scheduler on the Castor disk servers has improved file access rates for CMS and this, in turn, has improved the CPU efficiencies.
6) We did have a large backlog of file migrations to tape for Atlas through last week. On Friday we caught up.
Staffing:
1) No change since the last report. However, as that included significant changes I have reproduced it here:
2) Andrew has reduced his Tier-1 time down to 20% (from 50%)
3) Matthew Viljoen has left. (He was 60% Tier-1.)
4) Carmine Cioffi (one of the DBAs) will be leaving early September (He is 60% Tier-1.)
5) Plans for recruitment awaiting GridPP5.
Purchasing:
1) No change to status.
Note that RAL is closed both Monday and Tuesday of next week (31st Aug/1st Sep).
REVIEW OF ACTIONS
=================
567.1 DC to report back on CMS RAL CPU efficiency drop in April & May. Done.
568.1 DC to investigate the Open Compute Project and revert to the PMB with information about cost-savings, risks, and capital required. Partially done (DB to complete).
571.1 Maintenance – PG to do and DB to inform Tier-2 what we are pledging on their half – PG needs to answer so that DB can do this.
571.2 JC to speak to Frederic re GridPP35 talk. Ongoing
571.3 DC to talk to LUX-Zeplin re ReidPP35 talk. Ongoing
571.4 PG to talk to LSST re GridPP35 talk. Ongoing
571.5 on RAL, need to know who is speaking on CEPH. DK took note. Ongoing
571.6 PMB members get their quarterly reports in by the next PMB meeting in 2 weeks. Ongoing.
571.7 Gareth to update us on the network issues at RAL. Done
ACTIONS AS OF 24.08.15
======================
568.1 DC to investigate the Open Compute Project and revert to the PMB with information about cost-savings, risks, and capital required. Partially done (DB to complete).
571.1 Maintenance – PG to do and DB to inform Tier-2 what we are pledging on their half – PG needs to answer so that DB can do this.
571.2 JC to speak to Frederic re GridPP35 talk.
571.3 DC to talk to LUX-Zeplin re ReidPP35 talk.
571.4 PG to talk to LSST re GridPP35 talk.
571.5 on RAL, need to know who is speaking on CEPH. DK took note. Ongoing
571.6 PMB members get their quarterly reports in by the next PMB meeting in 2 weeks.
572.1 Regarding GridPP5: PG to liaise with institutes and check costs and corresponding RAL figures then do calculations based on staff costings – these must be done by Friday 28.8.15 or early w/c 31.8.15 at the latest. Also to check if Atlas propose any changes before sending (he is currently on 3-day course for ITIC).
572.2 DB to send note about consumables to STFC.
572.3 Regarding normalising the travel budget: DB will make a case noting all of the above and try to pare back travel costs as much as possible but taking account of the above issues. He will liaise with DK before sign-off with PMB.
572.4 Re email from Sarah Verth: DB will construct a generic paragraph for the introduction for 3 cases based on the Glasgow posts with a set of headings and circulate to others.
572.5 Re GridPP35: DB will email UK GridPP about the costs for the hotels and confirm allocated hotels.