GridPP PMB Minutes 373 F2F (15.01.10) ===================================== Present: David Britton (Chair), Sarah Pearce (remote), Andrew Sansum, Tony Doyle, Jeremy Coles, Pete Clarke, Steve Lloyd, Roger Jones, John Gordon, David Kelsey, Tony Cass (remote), Glenn Patrick, Robin Middleton, Dave Colling, Neil Geddes (Suzanne Scott, Minutes) Apologies: None Introduction & Goals for the Day ================================ DB advised that we had written an incomplete version at present of the GridPP4 proposal. By Monday v5 was required in order to circulate to the CB on Tuesday morning, as they were due to meet on Thursday. This meeting today was to settle all issues and get the document into a form to be signed off by the CB. It was noted that input from the CB may change things. The proposal was due to be submitted to STFC on 28th January. NG commented that the CB did not need a perfect document, they were only interested in the main issues. DB agreed that the focus would be on the main issues, eg: the Tier-2. DB advised that following that, there would only be one week of preparation before final submission to STFC on February 11th. Delay was impossible as there were future uncertainties. DK advised that it might even be a good tactic (if possible) to submit earlier before any announcement was made. Introductory sectons of GridPP4 bid - parts 3-5 =============================================== DB noted that the first page had minor issues; the Executive summary had to be re-visited. Regarding (3) Experiment Motivation, he hadn't received anything. PC noted that 'unprecedented' should be removed. JG commented that we know how much it costs because we're doing it already. NG noted 'extend' or 'wish to' would be better wording. DB asked how we would join all of these bits together? He needed clear annotated copies from everyone and he would amalgamate the changes - he would need to receive these within 48 hours. RJ noted that it was the tone overall that was important. (4) International Context ------------------------- DB noted it was preferable not to discuss the FTEs here - he would remove these. DB noted that in the big picture it was good to encourage opportunistic use. (4.1) EGI --------- DB asked if RM or JG had any changes to make? JG advised that 'other SSC' was not mentioned yet this is of interest to STFC, therefore we should mention it - it related to an astronomy and astrophysics area. NG believed that less might be more in this section. SP said that the reference to global tasks and international tasks might be a bit obscure. DB asked whether as a high-level summary the section worked ok? RM felt wLCG would be of more interest. JG noted that wLCG sat on top of EGEE and we have to go through EGEE to reach it. DB suggested we move the detail from here to a later section. SP thought it was good to introduce something about it here though, it was referenced later so a couple of paragraphs would be useful near the beginning. DB advised that this needed to dovetail with Section 13 - who would do this? TD or JG? ACTION 373.1 Re (4.1) EGI: JG to fix the paragraph and send TD the bits he had cut out. 373.2 RE (4.1) EGI: TD would respond by amending his section 13 accordingly. (5) UK Grid for Particle Physics -------------------------------- DB noted this section aimed to describe what the Grid is and what GridPP3 achieved. NG queried the CPU wording, it would be better to say 'deliver more effectively and efficiently', ie: demonstrating more effective use. DC asked whether the Project Map was useful? PC felt no, it was too small and not easy to understand. DB noted it would be larger in scale and it did give an idea of the different parts of the project. He agreed that it needed to be more readable. This would be revisited, with more text possibly. PC suggested it could be put in an appendix and could be read or not. DB asked SP to do a version where the wording in boxes was increased, the numbers removed, and the font increased. The navigation links were not required. SL would check the reference to the last OC paper, which seemed incorrect. ACTION 373.3 Re (5) UK Grid: SP to change the Project Map by increasing the size of the words within the boxes, remove the numbers, the fonts generally to be increased in size. Take out the navigation links. 373.4 Re (5) UK Grid: SP to check the reference to the last OC paper, which seemed incorrect. Re the Highlights of GridPP3: DB asked whether these details added anything? NG thought yes, it gets the message in early on and shows that things work. PC agreed noting this section contained positive 'it works' statements. DC suggested ending with a sentence saying we are as prepared as we possibly can be for the LHC. NG advised that a note was required about coping with ongoing users. DB added also there were now new challenges, 'new known unknowns'. AS thought that on p5, the OC quote used was now quite old. SP thought there was one about STEP'09 being good - it was agreed that this would be preferred. ACTION 373.5 Re (5) Highlights of GridPP3: SP to track down the more recent OC quote regarding the success of STEP'09. (5.1) High-level view of GridPP4 -------------------------------- DB advised that it was here we aimed to say where we were going, involving new challenges and new structure. PC thought that on p6, the problems noted seemed as if we hadn't thought of them - better wording re debugging was required. RJ suggested we can't solve the problem of distributed debugging. NG suggested re the level of error reporting we add 'close to'. NG would send his comments to DB re this section. NG, PC & RJ to send their comments to DB. ACTION 373.6 NG, PC & RJ to send comments to DB on section 5.1 - GridPP4 high-level view. Regarding the Work Packages, DB asked whether D & E should be combined? Should we re-visit this? DC advised no, keep them separate. This was agreed. It was agreed to change 'components' in the preceding paragraph to 'Work Packages'. NG noted that at the end of Section 5 we should add experiment requirements to lead on into the next section. Section 6 - Experiment Requirements =================================== (6.1) Experiment Computing Models (+ table) -------------------------------------------- DB noted that this had been an adapted legacy but was possibly too long? JG asked if the LHCb model had changed? DB highlighted terminology: 'scheduled and on demand analysis'. RJ didn't like 'local' as this was not accurate. DC suggested 'Group and chaotic user'. NG wanted something that was easier for the panel to understand. DB suggested that for ATLAS and CMS we would use 'Group and User' - 'Group' was scheduled analysis; and 'User' was 'on demand' analysis. This was agreed. Tony Medland had asked that we include ALICE. On p8 RJ noted that it may not be a good idea to retain the ILC, rather, 'design studies for future linear colliders'. RJ would provide text. ACTION 373.7 Re (6.1) RJ to provide DB with text for the ILC etc ('design studies for future linear colliders' or similar). (6.2) Global Resources ----------------------- DB advised that this was the only official table he had. PC asked what the message was to the reader? It was quite complex and not easily accessible. DB was happy to remove it. SL thought they should throw out these numbers. This was agreed. (6.3) Computing Resources -------------------------- DB noted the the resource table still needed work - he had cut and pasted this information from another spreadsheet. The year shows the growth - what numbers do we want in here? GP noted that the experiments understood these ok. For DC, RJ and GP, all numbers were still preliminary. RJ advised that in the past, the numbers have typically been reduced - the past was a good predictor. DB noted however that in the past things had been delayed, so we weren't in a strong position. SL advised that these were the numbers for now. PC thought that in the text above there was too much uncertainty and unknowns highlighted. We need to say strongly that these are the experiment plans and these are the numbers. TD noted we need to change the ALICE zeros and include figures. DB thought we could put them in and grey them out in 2013 and 2014, but at least to end 2011 we have exploitation. A contingency line should be included. ACTION 373.8 DB to add a contingency line to the table in section 6.3. Re UKQCD, DB noted that if they ask within the 10% line then that is OK, or else they can be included in 'others'. DB felt this would probably go away in any case. DB noted we needed to add a table on Tier-2 resources. SL suggested taking 12.5% and saying it is what we are asking for, and pledge 80%. This was agreed. DB advised that we could worry about the pledges in September. JC queried the wording of the last para in 6.2 re the UK being 'invited to estimate their likely contributions'. RJ noted it was better to say 'pledge the contributions'. (6.4) Network Requirements --------------------------- AS thought we didn't want a huge saga about network at the Tier-1. DB advised we needed to explain the backup - it was resilient but was part of the project. Did we need to mention costs being borne by the project? PC thought that by saying nothing it was like electricity - this was assumed it was to be paid for - it was best not to highlight. DB noted that above the line it would be funded, if not, we get no funding and then still have to pay. DB asked about the JANET connection and campus infrastructure? Was one note adequate? Was there anything we needed to flag here? There was a discussion on campus networks. DB asked if a data flow table was relevant? PC advised that if it was not needed, then don't use one. Could we get serious numbers really quickly to make this up- to-date? What was the point of the table? DB advised that it parallels the info and tables provided in other sections. He was happy to get rid of it but we did need a sequence of plots/slides to defend the network requirements if asked. It was agreed to drop Table 1. (6.5) Service Requirements --------------------------- DB suggested moving appendix B to here (relating to the MoU etc). NG thought we couldn't move all of it, it was too large. DB noted we can leave the old text as it is. AS advised that 24x7 response is now for the bulk of the services. (7) Experiment Requirements ---------------------------- (7.1) WP-A ========== SL thought we should explain what HEPSPEC is. JG to provide an explanatory sentence. ACTION 373.9 JG to provide an explanatory sentence for HEPSPEC for section 7.1. DB asked AS whether it was true that two out of the last five purchases had issues? Yes. It was agreed to leave it as is - this was to be expected and was built-in. There was discussion on the buffer level. NG thought we should remove 'try to'. DB would amend. ACTION 373.10 DB to remove 'try to' throughout. There was a discussion on power draw and capacity and costs involved in relation to racks etc. The meeting looked at the hardware costings showing disk plots and dates etc. The plot was a good estimate. Contingency would cover the error factor. AS was happy with the assumptions made and the safety factor included. DB noted that electricity costs were below the line but needed re- doing because they incorrectly included both the safety-factor built into the hardware provision and an additional safety-factor for the power. DB proposed re-doing this within time-constants. AS would do a scenario on electricity costs based on new input from DB. DB would send AS some new spend numbers. AS/DB would iterate on electricity costs. ACTION 373.11 AS to do a scenario on electricity costs for WP-A based on new input from DB. AS/DB to iterate. PC thought that the external view on the table would be that there are too many categories, especially for half posts. The machine room being distinct from the fabric team was not accurate? It was noted that as a reviewer you care less about internal details. AS advised that a number of categories could be moved into 'well-found lab' and the costs split within that. TD thought however that the table did map to people, so the table was clear - it didn't hide anything. DB noted we needed to map this to the half-page descriptions in Appendix A and have it in the same format, also use the same text from there. RJ noted there was the example of what was provided to the PPRP in relation to posts and the case details for them. DB suggested the table be developed and JC's text taken out. AS was keen to keep the table broken down in teams, and target the well-found lab in more places. PC thought the panel may say there are too many numbers in too many categories. DB noted we will make it clear that this is how it is actually organised. (7.2) WP-B ========== Tier-2 Manpower --------------- SL commented that the arguments were good but not in the right order. There was a better argument below the table, and the analysis requirements feed into the numbers. SL to re-arrange the text. ACTION 373.12 SL to re-arrange the text argument sections in 7.2. In the GridPP4 Roles table, RJ noted that 'local' suggested non-Grid, and 'simulation' implied that no analysis was being done. 'Group & Users' would be preferred. DB agreed that the role column would be changed to ATLAS Group & User Analysis, CMS Group & User Analysis, etc, and 'simulation' and 'local' would be removed. ACTION 373.13 DB to remove 'simulation' and 'local' from GridPP4 Roles table (p21), and insert instead: ATLAS Group & User Analysis, CMS Group & User Analysis, etc. There ensued a general discussion on wording. The distribution of the 6 (non ATLAS and CMS focused) FTE's was discussed in some detail and dovetailed with the ATLAS, CMS and potential EGI posts. DB thought that all sites should know they will be expected to contribute to global or international tasks. DB asked if there were other issues on the table re the balances? JC raised an issue about site leaders and having columns for site leaders and sysadmins. These would be added. RJ/AS would harmonise comments about the Tier-2 shrinking and the ATLAS sites expanding. ACTION 373.14 RJ/AS to harmonise statements on changes to the Tier-1 and Tier-2 sizes. Tier-2 Hardware --------------- SL asked if everyone was happy with the plots? DB had sent SL the spreadsheet - the numbers were not very different. SL to re-do his table accordingly and extend to 2015. ACTION 373.15 Re Tier-2 H'ware plots: SL to re-do his table according to the new numbers sent by DB, and extend to 2015. The last sentence needed work. DB asked if, in the table, we mix what we have with what we need? There ensued a discussion on CPU and disk replacement cycles. SL noted that the green line has the GridPP3 figures within it. RJ noted that the danger in the table was that 'available' didn't mean 'usable'. SL would amend the table, and it would need to be explained. ACTION 373.16 SL to amend the CPU and disk tables in the Tier-2 Hardware section, and an explanation of the table was required. (7.3) UK Tier Centre -------------------- DB noted this section was a legacy from GridPP3. (8) WP-C: Grid Deployment & Operations ====================================== DB thought this section was unbalanced and unclear, and seemed large in comparison with sect (8.1). JC advised that he had amended the text. DB advised that it needed strong definitions of posts/sites and had to be more focussed. JG said we should describe what we're doing. DB asked if JC wished him to re-draft it? It needed logical balance and progression - flow and balance needed to be changed. If JC had a new version, DB would critique it, asap as time was tight. JC to send by Monday and DB would iterate then. DK noted he would feed-in re security. ACTION 373.17 JC to send DB his revised draft of WP-C. (9) WP-D: Experiment-Specific Support ===================================== NG thought that this read strangely in comparison with the rest - the structure needed to be changed. TD commented that it looked like it stood on its own as a bid in its own right. JG suggested missing out the 1st sentence and half of the second, beginning at: 'This work package ..'. DB noted we could up the bid for a full FTE for both of these - the danger was putting in for extra that they may cut. There ensued a discussion on bidding and cost cutting in relation to extra half posts. It was agreed that the summary was better sited in the resource section. ACTION 373.18 GP to re-site the summary of WP-D within the resource section. DC noted that if we're given a budget explicity we should hit it. DK noted this could even be stated upfront. DB noted 2 x EGI posts and 2 x ROSCOE posts at some level may not be funded. It was agreed to keep the best case, remove the worst case, and move the contingency issue to the contingency section. 373.19 Re WP-D: GP to to keep the best case, remove the worst case, and move the contingency issue to the contingency section. There ensued a discussion on the documentation post: GP was trying to justify this post. DB advised that we had 2 x half posts with 2 x current postholders - what currently existed may not be realistic to propagate. DB asked what we wanted this post to do? GP thought it should be a traditional user support post. TD noted the context should be the complexity of the Grid. DB noted we were arguing for a person rather than a role - what was the role? DB classed it as 3rd line grid support? PC noted we couldn't re-factor it elsewhere to be part of a larger area. GP advised that where it was already situated was where experiment support was supposed to go. TD suggested expert consultancy? PC thought we could consolidate it elsewhere. DC commented that the Imperial 1/2 post was ok here. It was agreed to call it 1 x FTE in this area and footnote as 2 x 0.5 FTE which would be 'experiment-specific technical support including non- LHC experiments and assisting with technical middleware problems'. ACTION 373.20 Re WP-D: GP to note 1 x FTE in the area of technical support and documentation/trouble-shooting, footnoted as 2 x 0.5 FTE which would be 'experiment-specific technical support including non-LHC experiments and assisting with technical middleware problems'. (10) WP-E: Grid Support ======================= SP thought that the first sentence was very long. TD proposed 4 x FTEs in data management, replication and storage, which map onto existing roles. DB advised that the whole first paragraph was not needed? The wording 'support between sites' might be an issue. We need to say what Grid Support is and what we did in GridPP3, and what we need in GridPP4. DB noted that the posts should be 3.5 FTE + 0.5 FTE associated with data replication. JG noted that 3.5 out of 4.2 were data-related. DB asked whether we needed to tie some of the data staff more to the experiments? TD said no, three were embedded. RJ thought we could beef- up a post supporting generic technology in the Tier-2. NG/RJ to send comments to TD - this post needs to be broadened. We could mention LHCb and analysis at the Tier-1. ACTION 373.21 NG/RJ to send comments to TD regarding WP-E and broadening the experiment-related data post(s). On p30, DB thought that the itemised list was in doubt in comparison with objectives. DB thought mentioning achievements was good. A high-level objectives list + implicit tasks were not required - however a high-level justification was required. TD noted he could remove the detail and leave the general points. ACTION 373.22 TD to remove the detail and leave the general points on p30, providing a high-level justification rather than an objectives list. DB noted TD should focus on why, for users in the UK, is it important that we do this work? There was an understanding that we need to track and adapt to evolving environments - expert technical effort was required within the community. (11) WP-F: Management ====================== DB noted a move from terminology of d-Team to Operations Team. It was suggested that DB remove the arrows from the diagram. Also removing the first sentence of the following paragraph. ACTION 373.23 DB to remove the arrows from the diagram in WP-F. Also remove the first sentence of the following paragraph. Re the PMB table, it was possible that two people from Operations might come onto the PMB, strengthening communications with the Ops Team. DB suggested that a maximum of 10-12 people should comprise the PMB. DK noted that security should be included. AS thought the PMB table looked like a large management team. DB advised that the important point was that the number of people being funded was being reduced, but this should be re-visited. (12) WP-G: Impact Plan ======================= SP reported that STFC guidance said that we need to have such a plan and it is structured as per STFC criteria. Impae included KE and outreach. DB thought this section looked good - headings needed to be style-formatted. DK wasn't sure about the title. SP was happy to receive any further comments. ACTION 373.24 ALL: to send any comments to SP on the Impact section. (13) The Wider Grid/Computing Context ====================================== NG thought this was a rather narrow section that looked like we had been constrained to do this, then had left it. It was a hostage to fortune here - it looked like a nod in the direction rather than with purpose. We needed to show that our involvement was greater than this. DC suggested the possibility of running our own cloud in the future? DB suggested the approach that we were experts in the field - these were the current findings, but GridPP was looking for seed money to perhaps develop our own cloud. NG said it needed to read better as at the moment it sounded like an excuse not to deal with the subject. TD would amend it. There was a discussion on FTE effort. ACTION 373.25 TD to re-visit the Wider Grid Context section and amend tone and purpose of the section as per comments received. (14) Resource Request ====================== There was a discussion of disk and CPU plus Tier-2 hardware requirements in relation to DB's straw model on p40. ACTION 373.26 Re (14) Resource Request: DB to send SL the hardware costings to let SL check the Tier-2 sum. 373.27 Re (14) Resource Request: SP/DB need to converge on the remainder of the section. (14.11) Risks ------------- The PMB provided their '~top 3' risks: SL: 1. Tier-2 risk: loss of key experienced personnel 2. exposure to full infrastructure (electricity & cooling) costs 3. chaotic analysis model prevents full use of resources (sites cannot cope with on-demand user analysis) - mitigated by Data Management Post 4. insufficient manpower to operate core sites NG: 1. failure to deliver and report on NGI requirements and responsibilities 2. failure to support physicists at all sites - mitigated by S Burke being in post 3. failure to retain and recruit key technical staff due to perceived lack of support by STFC - mitigated by long-term funding of infrastructure projects by STFC DC (with CMS hat on): 1. loss of custodial data from the Tier-1 2. anything to cause substantial downtime at the Tier-1 3. Tier-2 not being able to cope with user analysis 4. supporting of physicists not at the 3 sites funded AS: 1. loss/damage to hardware 2. loss of custodial data 3. failure to meet MoU availability DK: 1. non-availability of services/compromise of data due to security incidents 2. reputation risk because of serious security incident(s) (ie: what could take the whole Grid down?) JC: 1. loss of key staff 2. major security incident bringing down the infrastructure 3. insufficient funding to meet the commitments at the Tier-2 centres RM: 1. travel: insufficient travel funds, compromising our engagement in wLCG 2. EGI/ROSCOE/Q not turning up with people 3. NGS funding in next round insufficient to keep NGI going at the level we require GP: 1. loss of Tier-1 experiment support staff 2. data storage problems stopping reconstruction 3. computing model changes and UK resources are insufficient to do what we intend RJ: 1. catastrophic loss of data 2. long-term problem with LFC 3. insufficient effort in primary group analysis sites 4. inability to deal with throughput of scaled-up infrastructure technologies JG: 1. hardware cost doesn't scale 2. GridPP/NGS tension/divergence 3. Tier-2 team tensioning - supporting hardware/experiment/operations = 3- way risk - mitigations are more people at the Tier-2 posts, tied to action (hypothications) 4. CASTOR PC: 1. failure of Tier-0 to Tier-1 link - short-term break risk is high but impact low 2. long-term failure - no mitigation 3. link to JANET sites: charge for JANET? change of financial model of JANET provision SP noted this could be included in unplanned infrastructure costs DB noted we had covered many things, his would include: 1. lack of adequate funding in future, including STFC changes/difficulty in getting the proposal through their system(s) DK noted that funding could change and requirements could change. DB also noted that exchange rates can cause problems. The other area of issue was: 2. divergence of the experiments - the Grid not working for real data ACTION 373.28 ALL: to send risks to SP by email, including mitigations, by tomorrow DB noted that sections that need work must be done by MONDAY, as the document had to be sent to the OC by TUESDAY. There was one week left between the OC and final submission of the first draft. Re the OC itself, DB had sent round a summary from Tony Medland - cuts may not happen to GridPP3 after all, which factors-in to GridPP4. SP noted she was waiting for AS to let her know about the move from FY11 to FY10. Tape media was a possibility. DB informed SP that he needed to know the cost to bridge the four x deputy Tier-2 Co-ords. DB advised that for the OC we would need a Project Summary report from everyone. DB would co-ordinate all actions from the OC. It was understood that the OC would not pay total attention to the GridPP4 Proposal as well as the standard OC report, plus all of the experiments' information - the aspiration was that there would be no confusion between the experiments and the GridPP4 proposal. There was no AOB and the meeting closed at 3.40 pm. All revised documents had to be sent to DB by Sunday. There would be no PMB on Monday. REVIEW OF ACTIONS ================= 354.2 JC to consult with site admins on a framework policy for releases, with a mechanism for escalation, plus a mechanism for monitoring. JC reported that the consultation happened. There were a few suggestions in the deployment team about how to progress in this area. It needs writing up and an implementation plan. 358.1 SP to work with the working group on the following issues in relation to GridPP/NGS convergence: 1. identify Institutes 2. identify manpower 3. decide who is bidding for what - a draft transition plan would be made available by the end of the year; GridPP4 requirements would also be considered. SP was waiting on the Working Group to reply to her. A meeting had been held before Christmas re a transition plan. SP was awaiting a skeleton outline plan from RM, allocating people to sections. 366.8 AS to confirm that the Tier-1 proposes to use Tape-based storage in the period 2011 - 2015. AS noted this depended on money costs. DB advised this related to long-term plans and power capacity. Physical footprint space? Alternatives? Early action on AS required. AS had sent tech questions round the team and would forward inputs when available. DC noted to the meeting that today was the 16th Nov - only 4 weeks remained until Imperial, by which time we needed to have made extensive progress. To be discussed at the F2F on Friday. 367.2 RM to fill-in the grey boxes on DB's UK NGI diagram of a minimal NGI, as to what NGS would be doing in the areas listed. 372.1 ALL: to bring their 'top 3' high-level risks for GridPP4 to the Friday F2F. Areas of key responsibility should be considered (rather than risks across the project overall). If everyone could also provide a 'top 12' overall, but 'top 3' specifically, that would be preferred. 372.2 AS to circulate a timeline for disk server installation completion and drives replacement. ACTIONS AS AT 15.01.10 ====================== 354.2 JC to consult with site admins on a framework policy for releases, with a mechanism for escalation, plus a mechanism for monitoring. JC reported that the consultation happened. There were a few suggestions in the deployment team about how to progress in this area. It needs writing up and an implementation plan. 358.1 SP to work with the working group on the following issues in relation to GridPP/NGS convergence: 1. identify Institutes 2. identify manpower 3. decide who is bidding for what - a draft transition plan would be made available by the end of the year; GridPP4 requirements would also be considered. SP was waiting on the Working Group to reply to her. A meeting had been held before Christmas re a transition plan. SP was awaiting a skeleton outline plan from RM, allocating people to sections. 366.8 AS to confirm that the Tier-1 proposes to use Tape-based storage in the period 2011 - 2015. AS noted this depended on money costs. DB advised this related to long-term plans and power capacity. Physical footprint space? Alternatives? Early action on AS required. AS had sent tech questions round the team and would forward inputs when available. DC noted to the meeting that today was the 16th Nov - only 4 weeks remained until Imperial, by which time we needed to have made extensive progress. To be discussed at the F2F on Friday. 367.2 RM to fill-in the grey boxes on DB's UK NGI diagram of a minimal NGI, as to what NGS would be doing in the areas listed. 372.1 ALL: to bring their 'top 3' high-level risks for GridPP4 to the Friday F2F. Areas of key responsibility should be considered (rather than risks across the project overall). If everyone could also provide a 'top 12' overall, but 'top 3' specifically, that would be preferred. 372.2 AS to circulate a timeline for disk server installation completion and drives replacement. 373.1 Re (4.1) EGI: JG to fix the paragraph and send TD the bits he had cut out. 373.2 RE (4.1) EGI: TD would respond by amending his section 13 accordingly. 373.3 Re (5) UK Grid: SP to change the Project Map by increasing the size of the words within the boxes, remove the numbers, the fonts generally to be increased in size. Take out the navigation links. 373.4 Re (5) UK Grid: SP to check the reference to the last OC paper, which seemed incorrect. 373.5 Re (5) Highlights of GridPP3: SP to track down the more recent OC quote regarding the success of STEP'09. 373.6 NG, PC & RJ to send comments to DB on section 5.1 - GridPP4 high-level view. 373.7 Re (6.1) RJ to provide DB with text for the ILC etc ('design studies for future linear colliders' or similar). 373.8 DB to add a contingency line to the table in section 6.3. 373.9 JG to provide an explanatory sentence for HEPSPEC for section 7.1. 373.10 DB to remove 'try to' throughout. 373.11 AS to do a scenario on electricity costs for WP-A based on new input from DB. AS/DB to iterate. 373.12 SL to re-arrange the text argument sections in 7.2. 373.13 DB to remove 'simulation' and 'local' from GridPP4 Roles table (p21), and insert instead: ATLAS Group & User Analysis, CMS Group & User Analysis, etc. 373.14 RJ/AS to harmonise statements on changes to the Tier-1 and Tier-2 sizes. 373.15 Re Tier-2 H'ware plots: SL to re-do his table according to the new numbers sent by DB, and extend to 2015. The last sentence needed work. 373.16 SL to amend the CPU and disk tables in the Tier-2 Hardware section, and an explanation of the table was required. 373.17 JC to send DB his revised draft of WP-C. 373.18 GP to re-site the summary of WP-D within the resource section. 373.19 Re WP-D: GP to to keep the best case, remove the worst case, and move the contingency issue to the contingency section. 373.20 Re WP-D: GP to note 1 x FTE in the area of technical support and documentation/trouble-shooting, footnoted as 2 x 0.5 FTE which would be 'experiment-specific technical support including non-LHC experiments and assisting with technical middleware problems'. 373.21 NG/RJ to send comments to TD regarding WP-E and broadening the experiment-related data post(s). 373.22 TD to remove the detail and leave the general points on p30, providing a high-level justification rather than an objectives list. 373.23 DB to remove the arrows from the diagram in WP-F. Also remove the first sentence of the following paragraph. 373.24 ALL: to send any comments to SP on the Impact section. 373.25 TD to re-visit the Wider Grid Context section and amend tone and purpose of the section as per comments received. 373.26 Re (14) Resource Request: DB to send SL the hardware costings to let SL check the Tier-2 sum. 373.27 Re (14) Resource Request: SP/DB need to converge on the remainder of the section. 373.28 ALL: to send risks to SP by email, including mitigations, by tomorrow INACTIVE CATEGORY ================= 359.4 JC to follow up dTeam actions from the DB, as follows: --------------------------- 05.02 dTeam to try and sort out CPU shares and priority resources, at Glasgow first (perhaps by raising the job priority in Panda). --------------------------- JC would check the situation with Graeme Stewart (who was currently on annual leave). JC followed up with Graeme and the other experiments. A test was started but this area has been deemed low priority and further progress is not expected for some time. ATLAS see no issues with contention. LHCb are not intending to pursue anything in this area. A CMS discussion has started but again it does not appear to be urgent. If the experiments are not pushing this internally then there is nothing for the deployment team to follow up! It was noted there was no priority in ATLAS at present, this will be pending for a while. Move to inactive as it is a long-term action. --------------------- THE NEXT PMB WOULD TAKE PLACE ON MONDAY 25 JANUARY AT 12:55 PM.