Atlas CAP Closeout Thanks to all the presenters for excellent and frank presentations Thanks to all...

15
Atlas CAP Closeout Atlas CAP Closeout Thanks to all the presenters Thanks to all the presenters for excellent and frank for excellent and frank presentations presentations

Transcript of Atlas CAP Closeout Thanks to all the presenters for excellent and frank presentations Thanks to all...

Atlas CAP CloseoutAtlas CAP Closeout

Thanks to all the presenters for Thanks to all the presenters for excellent and frank presentationsexcellent and frank presentations

Project OverallProject Overall

U.S. Atlas (and International U.S. Atlas (and International Atlas) have made considerable Atlas) have made considerable progress in the past 12 monthsprogress in the past 12 months

Budgets remain very tight, and Budgets remain very tight, and may jeopardize project goals, may jeopardize project goals, especially in the out yearsespecially in the out years

Management (1)Management (1)

Overall, the management of the Software & Overall, the management of the Software & Computing of the research program appears Computing of the research program appears healthy...healthy...

For FY2006-FY2007, an adequate management reserve For FY2006-FY2007, an adequate management reserve is being maintained. For FY2008, it is ephemeral, in is being maintained. For FY2008, it is ephemeral, in that the computing project deficits consume the full that the computing project deficits consume the full reserve (i.e. there is no reserve). Moreover, the Tier 1 reserve (i.e. there is no reserve). Moreover, the Tier 1 plans do not include the expected increase in overhead plans do not include the expected increase in overhead by 30%, which pushes the overhead well above the by 30%, which pushes the overhead well above the reserve. reserve.

RecommendationRecommendation: Make presentation of what : Make presentation of what is “unfunded” explicit.is “unfunded” explicit.

Minor point: Use of Access and Project does not yet Minor point: Use of Access and Project does not yet (apparently) include resource loading. These details (apparently) include resource loading. These details are being held at lower levels of project (software, are being held at lower levels of project (software, tier1, etc.).tier1, etc.).

Management (2)Management (2) (Formal) Change Control remains for the (Formal) Change Control remains for the

most part unimplemented, and is not being most part unimplemented, and is not being used as a management tool.used as a management tool.– The one change which exercised the mechanism The one change which exercised the mechanism

this year was an aggregate “omnibus” change, this year was an aggregate “omnibus” change, which is not a recommended approachwhich is not a recommended approach

– The move to PanDA should have invoked change The move to PanDA should have invoked change control under any reasonable definitions of triggerscontrol under any reasonable definitions of triggers

– The re-organization of the WBS should likewise The re-organization of the WBS should likewise have invoked change control, even if cost neutralhave invoked change control, even if cost neutral

– A 3 month milestone delay as a trigger is much too A 3 month milestone delay as a trigger is much too lax at this stage of the projectlax at this stage of the project

– ““Changes in scope / cost” as a trigger is too vague; Changes in scope / cost” as a trigger is too vague; recommend adding an FTE-month or $$$ recommend adding an FTE-month or $$$ qualificationqualification

RecommendationRecommendation: Utilize Change Control : Utilize Change Control more formally.more formally.

Software (1)Software (1)

Data analysis model has matured to the Data analysis model has matured to the level of having a concrete definition of level of having a concrete definition of the AOD the AOD (additional work needed is (additional work needed is recognized)recognized)– Further effort on optimizing read Further effort on optimizing read

performance of the AOD & ESD would be well performance of the AOD & ESD would be well spent nowspent now

– While the transient/persistent model shows While the transient/persistent model shows good initial gains, should study whether good initial gains, should study whether simplifying the persistent data structures will simplifying the persistent data structures will in the end result in higher performancein the end result in higher performance

Geometry effort impressive, facilitating a Geometry effort impressive, facilitating a nice event displaynice event display

Software (2)Software (2)

Responded well to problems in DC2 and Responded well to problems in DC2 and future scaling challenges by developing a future scaling challenges by developing a tool (PanDA) that minimized reliance on tool (PanDA) that minimized reliance on the weak links in the grid toolset. the weak links in the grid toolset. Change made at good point: early Change made at good point: early enough to do good job in hardening, not enough to do good job in hardening, not so early as to miss opportunity for so early as to miss opportunity for maturing middleware.maturing middleware.

Presentations need polish: PanDA might Presentations need polish: PanDA might be viewed as an adhoc, non-collaborative be viewed as an adhoc, non-collaborative effort. Need to emphasize thateffort. Need to emphasize that– at top end, interfaces to standard executorat top end, interfaces to standard executor– at the bottom end, interfaces to standard gridat the bottom end, interfaces to standard grid

Software (3)Software (3)

Commissioning strategy...Commissioning strategy...– Component level testing for CSC is Component level testing for CSC is

goodgood

Are the dates for milestones for Are the dates for milestones for components firm? Confirm they are components firm? Confirm they are early enough for integration testing.early enough for integration testing.

– Delaying a full scale data challenge Delaying a full scale data challenge to actual beam may represent a riskto actual beam may represent a risk

Software (3)Software (3)

Analysis Model needs additional Analysis Model needs additional work to show how pieces all work to show how pieces all connect togetherconnect together– will the AOD be used, and give good will the AOD be used, and give good

performance?performance?– role of tag database (vs trigger data) role of tag database (vs trigger data)

isn’t clear enough, relationship to isn’t clear enough, relationship to streamingstreaming

Software SupportSoftware Support

Plans for software support are considerably Plans for software support are considerably more concrete (good!)more concrete (good!)

How the 3 centers will be realized & staffed How the 3 centers will be realized & staffed remains vagueremains vague

RecommendationRecommendation: Identify people, charge : Identify people, charge them with the task, and get them to them with the task, and get them to organize their support activities to organize their support activities to accomplish the goals:accomplish the goals:– develop process that will motivate develop process that will motivate

support, and not overload those tasked support, and not overload those tasked with itwith it

– define / negotiate level of effortdefine / negotiate level of effort

FacilitiesFacilities

Substantial progress on Substantial progress on procurements and facility services procurements and facility services (e.g. dCache, grid interfaces) and (e.g. dCache, grid interfaces) and ramp-up of stafframp-up of staff

Wonderful news on WAN upgradesWonderful news on WAN upgrades Capacity and staff are on scheduleCapacity and staff are on schedule FY2006 will be another significant FY2006 will be another significant

year of increasesyear of increases

Facilities (2)Facilities (2)

While current scale of dCache is While current scale of dCache is impressive, current utilization is impressive, current utilization is modest and does not yet represent modest and does not yet represent a significant testa significant test

Size of resources utilized through Size of resources utilized through the grid interface averaged ~100 the grid interface averaged ~100 cpu’s, which is rather modest; cpu’s, which is rather modest; need to work on larger stress testsneed to work on larger stress tests

Facilities (3)Facilities (3)

BNL is enjoying significant benefit from BNL is enjoying significant benefit from co-location with RCFco-location with RCF– Need to document benefits, both to show Need to document benefits, both to show

the benefit to agencies, and to reveal the benefit to agencies, and to reveal potential risk if RCF goes awaypotential risk if RCF goes away

Budget does not include upcoming Budget does not include upcoming 30% increase in overhead.30% increase in overhead.

RecommendationRecommendation: include the : include the overhead in out-year planning.overhead in out-year planning.

Facilities (4)Facilities (4)

Long term planning for Tier 1 (space, Long term planning for Tier 1 (space, etc.) needs additional attentionetc.) needs additional attention

Current ESD size and simulation times Current ESD size and simulation times are considerably above “nominal”, and are considerably above “nominal”, and may result in need for additional may result in need for additional storage and compute resources, or a storage and compute resources, or a de-scoping of event rate.de-scoping of event rate.RecommendationRecommendation: Make sure this issue : Make sure this issue is being addressed at all levels within is being addressed at all levels within International AtlasInternational Atlas

Facilities (5)Facilities (5)

Procurement of 2008 resources Procurement of 2008 resources with fy2008 funds implies that with fy2008 funds implies that resources are not fully available resources are not fully available January 1.January 1.

RecommendationRecommendation::– Plans should be made to reflect the Plans should be made to reflect the

realities of procuring and realities of procuring and commissioning multiple petabytes of commissioning multiple petabytes of disk and ~1000 nodes (plan for disk and ~1000 nodes (plan for staged roll-out during Jan-April 2008)staged roll-out during Jan-April 2008)

GridsGrids

Formal involvement in OSG management Formal involvement in OSG management is on a good trackis on a good track– expressions of support from US Atlas expressions of support from US Atlas

management, involvement of Torremanagement, involvement of Torre There is some unease on both sides There is some unease on both sides

(OSG, US Atlas) about Atlas participation (OSG, US Atlas) about Atlas participation in OSGin OSGRecommendation: Find success oriented Recommendation: Find success oriented approach (find a win-win) for US Atlas approach (find a win-win) for US Atlas participation in OSG. OSG needs to participation in OSG. OSG needs to deliver a net positive impact on the US deliver a net positive impact on the US LHC experiments early in LHC operationsLHC experiments early in LHC operations