Title: LCG CR : APPLICATION AREA
1LCG CR APPLICATION AREA
P. Dauncey, M. Martinez-Perez, M. Mangano , S. D.
T. - referees
- BASED ON THE PARALLEL SESSION APPLICATION
AREA, Monday 25/6/06
2OUTLOOK
- APPLICATION AREA
- MANAGERIAL AND ORGANIZATIONAL ASPECTS
- MANPOWER STATUS
- STATUS AND ACHIEVEMENTS OF THE PROJECTS
- CONCLUSIVE REMARKS
3APPLICATION AREA
4APPLICATION AREA
- RECALL
- Deliver the common physics applications software
- Organized to ensure focus on real experiment
needs - Experiment-driven requirements and monitoring
- Open information flow and decision making
- Participation of experiment developers
- Frequent releases enabling iterative feedback
- Success defined by experiment validation
- Integration, evaluation, successful deployment
5AA Projects
REPORT INCLUDED IN THE OVERVIEW
- SPI Software process infrastructure (A.
Pfeiffer) - Software and development services external
libraries, savannah, software distribution,
support for build, test, QA, etc. - ROOT Core Libraries and Services (R. Brun)
- Foundation class libraries, math libraries,
framework services, dictionaries, scripting, GUI,
graphics, SEAL libraries, etc. - POOL Persistency Framework (D. Duellmann)
- Storage manager, file catalogs, event
collections, relational access layer, conditions
database, etc. - SIMU - Simulation project (G. Cosmo)
- Simulation framework, physics validation studies,
MC event generators, Garfield, participation in
Geant4, Fluka.
DEDICATED PRESENTATIONS
6MANAGERIAL AND ORGANIZATIONAL ASPECTS
7- Informal forum
- exchange of information between the AA projects
and experiments - Every two weeks
- Experiments participate in planning,
management, and architectural and technical
direction - Every two weeks
- Public minutes
- Project milestones, proposed by AA projects
- Discussed in the AF
- Added into quarterly reports for approval
- Quarterly reports
- To monitor the progress of the projects
- Reviewed by LCG Management Board
- Reviews
- Internal AA reviews
- LHCC Comprehensive Review
8MANPOWER, STATUS AND PERSPECTIVES
9Manpower Table
10Estimated Resource Needs
11Staffing Remarks
- Reduction of manpower as anticipated for 2006
- 42 FTE -gt 39 FTE (3FTE)
- But perhaps not in the projects or work packages
expected - The projects most affected by the reduction
- SPI 5 FTE -gt 2.5 FTE ( EGEE contribution IT)
- COOL 2.6 FTE -gt 1.7 FTE
- Project affected positively
- PROOF 2.6 FTE -gt 4.3 FTE (DOCT, MC Fellow)
- Major reduction is expected in 2008
- End LD contracts retirement. No solution yet
12STATUS OF THE PROJECTS
13SPI Software Project Infrastructure 1/3
- Andreas Pfeiffer replaced Alberto Aimar as SPI
project leader - coping with strong (50) reduction of resources
- Outsourcing some activities ?
- SPI services (quite a diversified collection)
- External Software (Common libraries from HEP, IT,
) - Savannah Project Portal
- enables project members to collect all the
information related to tracker items of their
choice - Build and Distribution
- LCG Software configuration management
- Builds and releases
- Software Distribution
- Quality Assurance Service
- Testing frameworks
- QA checklists and reports
- Documentation and Training
- Development of LCG policies, templates
- Code Documentation
14SPI Software Project Infrastructure 2/3
- Examples to quantify the activity
- Evolution of configurations
15SPI Software Project Infrastructure 3/3
- Examples to quantify the activity
- Status of Savannah
- 1944 users
- 1083 at last review 80
- 196 projects
- 122 at last review 60
16SIMU - SIMULATION PROJECT 1/4
- SIMULATION SUBPROJECTS
- Simulation framework
- Interface to multiple simulation engines (Geant4,
Fluka) and geometry models exchange (GDML), MC
truth handling, Python interfaces - Geant4 team participating
- Aligned with and responding to needs from LHC
experiments, physics validation, simulation
framework - Fluka team participating
- Framework integration, physics validation
- Garfield team participating
- Garfield package integration and support in LCG -
SPI - Simulation physics validation
- Assess adequacy of simulation and physics
environment for LHC and provide the feedback to
drive needed improvements - Generator services
- Generator libraries validation/test suite
development when needed (HepMC, etc.) common
event files
17SIMU - SIMULATION PROJECT 2/4
- GEANT4
- 2 new releases (Jan06, Jun06)
- Improvements in simulation (EM, calorimetry, back
scattering, etc.) - Improved robustness
- Strong collaboration with Physics Validation
subprojects - FLUKA
- new realease (Sep06) 2006-3
- several physics improvements, f.i.
- Generation of primary ionisation events (request
by ALICE) - New high-energy hadronic generator
- Improvements in the evaporation/fission models
- First implementation of photon-muon pair
production - GARFIELD
- interfaces and porting
- computational methods
18SIMU - SIMULATION PROJECT 3/4
- PHYSICS VALIDATION
- Background radiation studies
- Goal comparison of Fluka and Geant4 predictions
- Radiation studies in LHCb, CMS
- TARC benchmark test, in particular for n
- Simple benchmarks
- inclusive p production in p,K,p,p interactions
on Mg, Ag, Au, at 100 and 320 (p-) GeV/c
results ready! - Test beam setups
- ATLAS EM barrel test beam 2002
- CMS HCAL test beam 2004
- ATLAS HEC test beam (Geant4 release validation)
- ATLAS barrel combined test beam 2004
- ATLAS TileCal test beam 2002
- investigations to improve hadronic shower shapes
in the simulation - relation between simulation developers and
experiments has improved
19SIMU - SIMULATION PROJECT 4/4
- GENSER - GENERATOR SERVICES subproject
- new release last summer 1.4.0
- LHC experiment requests collected in LCG
Generator monthly meetings - critics expressed by MC authors in recent MC4LHC
workshop Jul06 - Procedural mistakes in releasing and fixing bugs
- Packaging issues for some MC generators
- More testing needed for core generators
- Control of feedback and support requests from
experiments - experiments want GENSER to continue its mandate
- ? new project leader Witold Pokorski (PH/SFT)
- field of major progress
- PYTHIA 8 (stand-alone C, with 3 year road-map,
Sep 2004 - Sep 2007) - MCDB (MC event Data Base) ? Ready to use !
- Generators framework General interface to any
generator provided by GENSER (Production
Validation) - GDML (Geometry Description Markup Language)
- Extended to support all solids / constructs of
Geant4 - Added support for optical surfaces and material
properties
20ROOT PROJECT 1/3
- ORGANIZATION
- BASE Fons Rademakers, Bertrand Bellenot(also
SPI) - I/O Philippe Canal, Paul Russo(FNAL), Markus
Frank(lt10) - DICT Philippe Canal, Axel Naumann, Stefan Roiser
- MATH Lorenzo Moneta, Anna Kreshouk
- GUI Ilka Antcheva, Bertrand Bellenot
- GRAPHICS Olivier Couet
- GEOM Andrei Gheata, Mihaela Gheata (Alice)
- PROOF Fons, Maarten Ballintijn(MIT), Gerri
Ganis, Bertrand , Leandro Franco, Ian
Iwaszkiewicz, Andreas Peters(arda) - SEAL Lorenzo
- MAIN RECENT EVENTS
- ROOT Workshop at CERN October 2005
- 3 PRO versions in the last y, next Dec06
- DEV versions, next Oct06
21ROOT PROJECT 2/3
- RECENT ACHIEVEMENTS (a selection)
- New MonaLisa monitoring plug-in
- New Users Guide
- More friendly user interfaces
- Implemented fast merging of Trees without
unzipping the branch buffers - storing and restoring of TTree to and from an SQL
database - Implemented a TTreeCache with a huge improvement
when accessing remote files - ROOT/CINT/Reflex integration still many unknowns
- first version in next PRO version
- several progresses in the mathematical packages
- vectors, matrices, math. Function, numerical
calculations - Fitting and minimization (new C MINUIT)
- Improved random numbers
- Histogram comparison
- Multivariable analysis
- Important progresses in GUI sector
- GEOMETRY PACKAGE
- Interfaces to Geant3 and Fluka validated
- Interface to Geant4 available in the coming weeks
22Example of TTreeCache improvement
- The file is on a CERN machine connected to the
CERN LAN at at 100MB/s. - The client A is on the same machine as the file
(local read) - The client B is on a CERN LAN connected at 100
Mbits/s with a network latency of 0.3
milliseconds (P IV 3 Ghz). - The client C is on a CERN Wireless network
connected at 10 Mbits/s with a network latency of
2 milliseconds (Mac Intel Coreduo 2Ghz). - The client D is in Orsay (LAN 100 Mbits/s)
connected to CERN via a WAN with a bandwith of 1
Gbits/s and a network latency of 11 milliseconds
(P IV 3 Ghz). - The client E is in Amsterdam (LAN 100 Mbits/s)
connected to CERN via a WAN with a bandwith of 10
Gbits/s and a network latency of 22 milliseconds
(AMD64 280). - The client F is connected via ADSL with a
bandwith of 8Mbits/s and a latency of 70
milliseconds (Mac Intel Coreduo 2Ghz). - The client G is connected via a 10Gbits/s to a
CERN machine via Caltech latency 240 ms. - The times reported in the table are realtime
seconds
client latency(ms) cachesize0 cachesize64KB
cachesize10MB A 0.0 3.4
3.4 3.4 B 0.3 22.0
6.0 4.0 C 2.0
11.6 5.6 4.9 D 11.0
124.7 12.3 9.0 E
22.0 230.9 11.7 8.4
F 72.0 743.7 48.3
28.0 G 240.0 gt1800 125.4
9.9
One query to a 280 MB Tree I/O 6.6 MB
23ROOT PROJECT 3/3
- PROOF
- an extension of ROOT to run on a distributed,
heterogeneous system - 4.1 FTE dedicated
24PERSISTENCY FRAMEWORK 1/3
- 3 Software Packages
- CORAL
- Abstraction of relational database access for
Oracle, MySQL, SQLight and FroNTier - POOL
- C object persistency (via Root or Databases)
and navigation (via catalogs and collections) - COOL
- Management of versioned conditions time series
- Layered and complementing each other
- COOL and POOL use CORAL to access databases
- Experiments use COOL to reference conditions
objects stored in POOL
25PERSISTENCY FRAMEWORK 2/3
- RECENT ACHIEVEMENTS AND STATUS (a selection)
- CORAL
- Integration with LCG 3D services
- Database service at T0 and T1s
- Replica management via existing LFC service
- now optionally hosted in an LFC catalog (before
implementation based on XML file) - Next frontier consolidation
- POOL
- Consolidation and maintenance phase
- Significant work to absorb changes of the core
infrastructure (f.i. SEAL?ROOT) - Good news
- consolidation on fewer grid catalogs
(CLARIFICATION WITHIN THE EXPERIMENTS!) - LFC seems to be established as baseline
- CMS also developed uses Trivial File Catalog
- POOL did not have to invest in new catalog
integration - New functionality added dictionary auto-loading
- Enables the loading-on-demand of the required
dictionary libraries at run time
26PERSISTENCY FRAMEWORK 3/3
- RECENT ACHIEVEMENTS AND STATUS, cont.
- COOL
- 21 releases since the beginning (Nov2004)
- recently 1 FTE less !
- In general
- need to ensure continuity with expert manpower
- PF components rely on SEAL component model
- Propose to setup a table of SEAL components and
plan their evolution
27MORE ABOUT AA STATUS
- Release frequency
- Typically 2-3 production releases/y
- Bug fix releases - as many as needed to minimize
TTU (time to user) - Minimizing TTU a major goal for next year
- AA interfaces to Middleware and Fabric
- Many contact points
- Problem observed due to lack of coordination
- Milestones since last CR
28CONCLUSIVE REMARKS
29AA CONCLUSIVE REMARKS 1/2
- abundant amount of work
- certainly important progress and achievements
- the level of progress is not the same in the
various projects and subprojects - the intrinsic fragmentary nature of part of the
projects ? managerial difficulties - local manpower limitations (f.i. Software Process
infrastructure) - the Simulation project
- need to overcome the present difficulties in
interfacing some Monte Carlo generators to the
LCG simulation infrastructure - ROOT project
- properly managed
- appropriate manpower resources
- important achievements, in particular
consolidation, fast access to data - Merging of SEAL (Shared Environment for
Applications at LHC) progressing successfully - important progress of PROOF, powerful tool to
extend ROOT to run on a distributed,
heterogeneous system - Alice, CMS and LHCb are expressing interest in
using of PROOF - clear decisions by the experiments needed
30AA CONCLUSIVE REMARKS 2/2
- persistency framework project
- key ingredient for LHC computing
- difficulty to access the progress level, limited
help from the material presented - important effort by AA to keep the link with the
experiments and the users strong and effective - Architects Forum, AA Meetings every 2 weeks
- LCG Generator monthly meetings
- Savannah portal
- manpower
- present level globally very near to the needs
- some reassignment can cure the limitations
affecting individual projects - possible manpower crisis in 2008 (retirements and
contract ends) - appropriate action be taken in2007 to guarantee
adequate manpower level in 2008 and beyond