Скачать презентацию Introduction and Overview of the ALICE Computing Model Скачать презентацию Introduction and Overview of the ALICE Computing Model

2511ec9e3e631408abcab5caae37fa96.ppt

  • Количество слайдов: 48

Introduction and Overview of the ALICE Computing Model F. Carminati ALICE Computing Model Workshop Introduction and Overview of the ALICE Computing Model F. Carminati ALICE Computing Model Workshop December 9 -10, 2004 December 9, 2004 Computing Model Workshop

Objective of the meeting • Present the current status of the ALICE • • Objective of the meeting • Present the current status of the ALICE • • • computing model Receive feedback from the Collaboration Receive endorsement for the draft to be presented to the LHCC review committee Start the process that will bring to the Computing TDR December 9, 2004 Computing Model Workshop 2

Timeline • December 15: the draft computing model and the • projected needs are Timeline • December 15: the draft computing model and the • projected needs are presented to the LHCC review committee January 17 -19 LHCC review with sessions devoted to each of the experiments and a closeout session § Monday, 17 January : ATLAS (a. m. ), CMS (p. m. ) § Tuesday, 18 January: ALICE (a. m. ) LHCb (p. m. ) § Wednesday, 19 January: Closed Session (a. m. ) December 9, 2004 Computing Model Workshop 3

Computing TDR’s • LCG TDR § § Next meeting first half 2005 First draft Computing TDR’s • LCG TDR § § Next meeting first half 2005 First draft 11 April, good copy 9 May 15 June final TDR to LHCC (LHCC mtg. 29 -30 June) 3 June ready for approval by PEB on 7 June • ALICE Computing TDR § Early draft given to LHCC on December 15 § Draft presented and distributed to the Collaboration during the ALICE/offline week in February § Final discussion and approval during the ALICE/offline week beginning of June December 9, 2004 Computing Model Workshop 4

Computing Mo. U • Distributed to the Collaboration management to • • • obtain Computing Mo. U • Distributed to the Collaboration management to • • • obtain feedback on October 1 Coupled with the LHCC review in February Provide the C-RRB with documents that can be finalised and approved at its April 2005 meeting Subsequently distributed for signature December 9, 2004 Computing Model Workshop 5

Mandate of the February LHCC review • In the context of the preparation of Mandate of the February LHCC review • In the context of the preparation of the Computing Mo. Us and TDRs, • the LHC experiments have come forward with estimated computing capacity requirements in terms of disks, tapes, CPUs and networks for the Tier-0, Tier-1 and Tier-2 centres. The numbers vary in many cases (mostly upwards) from those submitted to the LHC Computing Review in 2001 […] it is felt to be desirable at this stage to seek an informed, independent view on the reasonableness of the present estimates. […] the task of this Review is thus to examine critically, in close discussion with the computing managements of the experiments, the current estimates and report on their validity in the light of the presently understood characteristics of the LHC experimental programme. The exercise will therefore not be a review of the underlying computing architecture. December 9, 2004 Computing Model Workshop 6

Membership • Chairman: J. Engelen - CERN Chief Scientific Officer • Representatives from the Membership • Chairman: J. Engelen - CERN Chief Scientific Officer • Representatives from the LHCC: F. Forti, P. Mc. Bride, • • T. Wyatt External: E. Blucher (Univ. Chicago), N. N. LHCC Chairman and Secretary: S. Bertolucci, E. Tsesmelis PH Department: J. -J. Blaising, D. Schlatter IT Department: J. Knobloch, L. Roberston, W. von Rueden December 9, 2004 Computing Model Workshop 7

Elements of the computing model • • • From detector to Raw data (see Elements of the computing model • • • From detector to Raw data (see P. Vande Vyvre’s talk) Framework & software management Simulation Reconstruction Condition infrastructure Analysis Grid Middleware & distributed computing environment Project management & planning From RAW data to physics analysis (see Y. Schutz’s talk) December 9, 2004 Computing Model Workshop 8

Framework • Ali. Root in development since 1998 § Entirely based on ROOT § Framework • Ali. Root in development since 1998 § Entirely based on ROOT § Used already for the detector TDR’s • Two packages to install (ROOT and Ali. Root) § Plus transport MC’s • Ported on several architectures (Linux IA 32 and IA 64, • Mac OSX, Digital True 64, Sun. OS…) Distributed development § Over 50 developers and a single cvs repository • Tight integration with DAQ (data recorder) and HLT (same codebase) December 9, 2004 Computing Model Workshop 9

Ali. Root layout G 3 G 4 FLUKA ISAJET Ali. En/g. Lite Ali. Root Ali. Root layout G 3 G 4 FLUKA ISAJET Ali. En/g. Lite Ali. Root Virtual MC Ali. Reconstruction HIJING EVGEN Ali. Simulation HBTAN PMD EMCAL STRUCT CRT TRD START PYTHIA 6 STEER ITS PHOS FMD TOF MUON MEVSIM PDF ZDC TPC RICH HBTP RALICE ESD Ali. Analysis ROOT December 9, 2004 Computing Model Workshop 10

Software management • Regular release schedule § Major release every six months, minor release Software management • Regular release schedule § Major release every six months, minor release (tag) every month • Emphasis on delivering production code § Corrections, protections, code cleaning, geometry • Nightly produced UML diagrams, code listing, coding rule violations, build and tests , single repository with all the code § No version management software (we have only two packages!) • Advanced code tools under development with IRST/Italy § Aspect oriented programming § Smell detection § Automated testing December 9, 2004 Computing Model Workshop 11

Simulation • Simulation performed with Geant 3 till now • Virtual Monte. Carlo interface Simulation • Simulation performed with Geant 3 till now • Virtual Monte. Carlo interface separates the ALICE code • • from the Monte. Carlo used New geometrical modeller scheduled to enter production at the beginning of 2005 Interface with FLUKA finishing validation The Physics Data Challenge 2005 will be performed with FLUKA Interface with Geant 4 ready to be implemented § Second half 2005 (? ) • Testbeam validation activity started December 9, 2004 Computing Model Workshop 12

The Virtual MC Geant 4_mc. tar. gz includes the TVirtual. MC <--> Geant 4 The Virtual MC Geant 4_mc. tar. gz includes the TVirtual. MC <--> Geant 4 interface classes Geant 3. tar. gz includes an upgraded Geant 3 with a C++ interface G 4 VMC G 3 transport G 4 transport FLUKA User Code G 3 FLUKA transport Reconstruction Geometrical Modeller Visualisation Generators December 9, 2004 Computing Model Workshop 13

HMPID: 5 Ge. V Pions Geant 3 December 9, 2004 Computing Model Workshop FLUKA HMPID: 5 Ge. V Pions Geant 3 December 9, 2004 Computing Model Workshop FLUKA 14

TGeo modeller December 9, 2004 Computing Model Workshop 15 TGeo modeller December 9, 2004 Computing Model Workshop 15

Reconstruction strategy • Main challenge - Reconstruction in the high flux • environments (occupancy Reconstruction strategy • Main challenge - Reconstruction in the high flux • environments (occupancy in the TPC detector up to 40%) requires a new approach to tracking Basic principle – Maximum information principle § use everything you can, you will get the best • Algorithms and data structures optimized for fast access and usage of all relevant information § Localize relevant information § Keep this information until it is needed December 9, 2004 Computing Model Workshop 16

Tracking strategy – Primary tracks • Iterative process § Forward propagation towards to the Tracking strategy – Primary tracks • Iterative process § Forward propagation towards to the vertex –TPC-ITS § Back propagation – ITS-TPC-TRD-TOF § Refit inward TOFTRD-TPC-ITS TOF TRD TPC ITS • Continuous seeding –track segment finding in all detectors December 9, 2004 Computing Model Workshop 17

Sources of information • • spatial characteristic of a track and sets of tracks Sources of information • • spatial characteristic of a track and sets of tracks § § § § px, py, pz, y, z parameters and covariance chi 2 number of points on the track number of shared clusters on the track overlaps between tracks DCA for V 0 s, Kinks and Cascades … d. Edx § mean, sigma, number of points, number of shared points… reliability TOF of a track and sets of tracks derived variables § § Mass Causality - Probability that particle “ really exists” in some space interval (used for causality cuts) • Based on clusters occurrence, and chi 2 before – after vertex § Invariant mass § Pointing angle of neutral mother particle § … December 9, 2004 Computing Model Workshop 18

ITS tracking • Follow the TPC seeds into a tree of track hypotheses connecting ITS tracking • Follow the TPC seeds into a tree of track hypotheses connecting reconstructed clusters § track in dead zone § missing clusters (dead or noisy channels, clusters below threshold) § secondary tracks not cross ITS layer as function of impact parameter in z and r-φ § probability of the cluster to be shared as a function of the cluster shape § restricted amount of tracks kept for further parallel tracking procedure § for secondary tracks also short best tracks kept, for further V 0 study • Best track is registered to all the clusters which belong to that track • Overlap between the best track and all other tracks is calculated, and if above threshold, χ2 of the pair of tracks is calculated December 9, 2004 Computing Model Workshop 19

ITS - Parallel tracking (2) Conflict ! Best track 1 Best track 2 • ITS - Parallel tracking (2) Conflict ! Best track 1 Best track 2 • double loop over all possible pair of branches • weighted χ2 of two tracks calculated § effective probability of cluster sharing and for secondary particles the probability not to cross given layer taken into account December 9, 2004 Computing Model Workshop 20

Results – Tracking efficiency (TPC) • PIV 3 GHz – (d. N/dy – 6000) Results – Tracking efficiency (TPC) • PIV 3 GHz – (d. N/dy – 6000) § TPC tracking - ~ 40 s § TPC kink finder ~ 10 s § ITS tracking ~ 40 s § TRD tracking ~ 200 s December 9, 2004 Computing Model Workshop 21

Kink finder efficiency • Efficiency for Kaons as a function of decay radius • Kink finder efficiency • Efficiency for Kaons as a function of decay radius • Left side – low multiplicity (d. N/dy~2000) – 2000 Kaons • Right side – same events merged with central event (d. N/dy~8000) December 9, 2004 Computing Model Workshop 22

PID combined over several detectors Probability to be a particle of i-type (i = PID combined over several detectors Probability to be a particle of i-type (i = e, m, p, K, p, … ), if we observe a vector S= {s. ITS, s. TPC, s. TOF, …} of PID signals: is the combined response function. Ci are the same as in the single detector case (or even something reasonably arbitrary like Ce~0. 1, Cm~0. 1, Cp~7, CK~1, …) The functions R(S|i) are not necessarily “formulas” (can be “procedures”). Some other effects (like mis-measurements) can be accounted for. December 9, 2004 Computing Model Workshop 23

PID combined over ITS, TPC and TOF (Kaons) Selection : ITS & TPC & PID combined over ITS, TPC and TOF (Kaons) Selection : ITS & TPC & TOF (central Pb. Pb HIJING events) ITS TPC TOF ITS & TPC & TOF Efficiency Contamination Efficiency of the combined PID is higher (or equal) and the contamination is lower (or equal) than the ones given by any of the detectors stand-alone. December 9, 2004 Computing Model Workshop 24

HLT Monitoring CASTOR Root file Ali. En ESD Monitoring Histograms Aliroot Simulation Online Monitoring HLT Monitoring CASTOR Root file Ali. En ESD Monitoring Histograms Aliroot Simulation Online Monitoring Digits Raw Data December 9, 2004 Computing Model Workshop LDC LDC alimdc Event builder GDC 25

Condition Data. Bases • Information source stored in heterogeneous • • databases A program Condition Data. Bases • Information source stored in heterogeneous • • databases A program periodically polls all sources and creates ROOT file with condition information These files are published on the Grid Distribution of the files is done by the Grid DMS Files are identified via DMS metadata December 9, 2004 Computing Model Workshop 26

External relations and DB connectivity • Relations between DBs not final not all shown External relations and DB connectivity • Relations between DBs not final not all shown Physics data files ECS API DAQ API Trigger From URs: Source, volume, granularity, update frequency, access pattern, runtime environment and dependencies calibration procedures API DCS API DCDB API HLT calibration files Calibration classes Ali. En/GLite: API API – Application Program Interface December 9, 2004 metadata file store Ali. Root Call for UR to come!! Computing Model Workshop 27

Development of Analysis • Analysis Object Data designed to be analysis oriented § Contains Development of Analysis • Analysis Object Data designed to be analysis oriented § Contains data needed to analysis only § Designed for efficiency of the analysis • Analysis à la PAW § ROOT + at most a small • Work on the infrastructure done by the ARDA project • Batch analysis infrastructure § Prototype end 2004 • Interactive analysis infrastructure § Demonstration end 2004 • Physics working groups here just starting December 9, 2004 Computing Model Workshop 28

Site A Site B PROOF SLAVE SERVERS PROOF SLAVE LCG SERVERS Proofd Rootd Forward Site A Site B PROOF SLAVE SERVERS PROOF SLAVE LCG SERVERS Proofd Rootd Forward Proxy New Elements Optional Site Gateway Only outgoing connectivity Site Slave ports mirrored on Master host Proofd Startup Grid Service Interfaces Slave Registration/ Booking- DB TGrid UI/Queue UI Master Setup PROOF Steer Grid Access Control Service Grid/Root Authentication Grid File/Metadata Catalogue “Standard” Proof Session Master Booking Request with logical file names Client retrieves list of logical file (LFN + MSN) Grid-Middleware independend PROOF Setup December 9, 2004 PROOF Master Computing Model Workshop PROOF Client 29

December 9, 2004 Computing Model Workshop 30 December 9, 2004 Computing Model Workshop 30

The ALICE Grid (Ali. En) There are millions lines of code in OS dealing The ALICE Grid (Ali. En) There are millions lines of code in OS dealing with GRID issues Why not using them to build the minimal GRID that does the job? l l l Fast development of a prototype, can restart from scratch etc Hundreds of users and developers g. Lite Immediate adoption of emerging standards Ali. En by ALICE (5% of code developed, 95% imported) 2001 2002 2003 2004 Start 2005 10% Data Challenge (analysis) Physics Performance Report (mixing & reconstruction) First production (distributed simulation) Functionality + Simulation December 9, 2004 Interoperability + Reconstruction Performance, Scalability, Standards + Analysis Computing Model Workshop 31

Why Physics Data Challenges? • We need simulated events to exercise physics • • Why Physics Data Challenges? • We need simulated events to exercise physics • • • reconstruction and analysis We need to exercise the code and the computing infrastructure to define the parameters of the computing model We need a serious evaluation of the Grid infrastructure We need to exercise the collaboration readiness to take and analyse data December 9, 2004 Computing Model Workshop 32

PDC 04 schema Ali. En job control Production of RAW Shipment of RAW to PDC 04 schema Ali. En job control Production of RAW Shipment of RAW to CERN Reconstruction of RAW in all T 1’s Analysis E H Data transfer T N CERN O L !! A IT RID O G D Tier 2 December 9, 2004 Tier 1 Computing Model Workshop Tier 1 Tier 2 33

Merging Signal-free event December 9, 2004 Computing Model Workshop Mixed signal 34 Merging Signal-free event December 9, 2004 Computing Model Workshop Mixed signal 34

Phase II (started 1/07) – statistics • In addition to phase I § Distributed Phase II (started 1/07) – statistics • In addition to phase I § Distributed production of signal events and merging with phase I events § Network and file transfer tools stress § Storage at remote SEs and stability (crucial for phase III) • Conditions, jobs …: § § 110 conditions total 1 million jobs 10 TB produced data 200 TB transferred from CERN § 500 MSI 2 k hours CPU • End by 30 September December 9, 2004 Computing Model Workshop 35

 • Structure of event production in phase II Central servers Master job submission, • Structure of event production in phase II Central servers Master job submission, Job Optimizer (N sub-jobs), RB, File catalogue, processes monitoring and control, SE… Register in Ali. En FC: LCG SE: LCG LFN = Ali. En PFN Sub-jobs Ali. En-LCG interface Sub-jobs Storage Underlying event input files CERN CASTOR: underlying events RB CEs Job processing Output files Storage CEs CERN CASTOR: backup copy Job processing Output files zip archive of output files Local SEs Primary copy December 9, 2004 Local SEs Primary copy Computing Model Workshop File catalogue edg(lcg) copy®ister 36

UI shell December 9, 2004 Computing Model Workshop end middleware to application 37 end UI shell December 9, 2004 Computing Model Workshop end middleware to application 37 end

 • Structure analysis in phase 3 Central servers Master job submission, Job Optimizer • Structure analysis in phase 3 Central servers Master job submission, Job Optimizer (N sub-jobs), RB, File catalogue, processes monitoring and control, SE… Sub-jobs lfn 1 lfn 2 lfn 3 lfn 4 lfn 5 lfn 6 lfn 7 lfn 8 Ali. En-LCG interface PFN = (LCG SE: ) LCG LFN PFN = Ali. En PFN RB Job splitter Get PFN’s File catalogue Metadata Query LFN’s CEs Job processing Input files CEs User query Job processing Input files Local SEs Primary copy December 9, 2004 Local SEs File catalogue Primary copy Computing Model Workshop 38

Phase III - Execution Strategy • Very labour intensive § The status of LCG Phase III - Execution Strategy • Very labour intensive § The status of LCG DMS is not brilliant • Does not “leverage” the (excellent!) work done in ARDA § So… why not doing it with g. Lite? • Advantages § Uniform configuration: g. Lite on EGEE/LCG-managed sites & on ALICE-managed sites § If we have to go that way, the sooner the better § Ali. En is anyway “frozen” as all the developers are working on g. Lite/ARDA • Disadvantages § It may introduce a delay with respect to the use of the present – available – Ali. En/LCG configuration § But we believe it will pay off in the medium term • PEB accepted to provide us with limited support for this exercise § Provided it does not hinder the EGEE release plans December 9, 2004 Computing Model Workshop 39

New phase III - Layout g. Lite/A CE/SE User Query Server lfn 1 lfn New phase III - Layout g. Lite/A CE/SE User Query Server lfn 1 lfn 2 lfn 3 lfn 4 lfn 5 lfn 6 lfn 7 lfn 8 g. Lite/E CE/SE g. Lite/L CE/SE g. Lite/A CE/SE Catalog December 9, 2004 Computing Model Workshop 40

DAQ Management Board HLT US Grid coord. EU Grid coord. Software Projects Regional Tiers DAQ Management Board HLT US Grid coord. EU Grid coord. Software Projects Regional Tiers Offline Board Chair: Comp Coord Detector Projects LCG SC 2, PEB, GDB, POB Offline Coord. (Deputy PL) International Computing Board Core Computing and Software Production Environment Coord. • Production environment (simulation, reconstruction & analysis) • Distributed computing environment • Database organisation Framework & Infrastructure Coord. • Framework development (simulation, reconstruction & analysis) • Persistency technology • Computing data challenges • Industrial joint projects • Tech. Tracking • Documentation December 9, 2004 Simulation Coord. • • Detector Simulation Physics simulation Physics validation GEANT 4 integration FLUKA integration Radiation Studies Geometrical modeler Reconstruction & Physics Soft Coord. • Tracking • Detector reconstruction • Global reconstruction • Analysis tools • Analysis algorithms • Physics data challenges • Calibration & alignment algorithms Computing Model Workshop Offline Coordination • Resource planning • Relation with funding agencies • Relations with C-RRB 41

Core Computing Staffing December 9, 2004 Computing Model Workshop 42 Core Computing Staffing December 9, 2004 Computing Model Workshop 42

Physics Analysis Software Central Support Core Software Detector proj Physics WGs Subdetector Software M&O Physics Analysis Software Central Support Core Software Detector proj Physics WGs Subdetector Software M&O A Offline Coordination Comp proj Funding Infrastructure & Services Core Computing December 9, 2004 Computing Workshop Computing Model Project 43

CERN Core Offline 20 ~15 Extended Core Offline activities in the other projects December CERN Core Offline 20 ~15 Extended Core Offline activities in the other projects December 9, 2004 Computing Model Workshop 10 ~7 100 ~500? 44

Time Projection Chamber (TPC) M. Kowalski, M. Ivanov Core Offline P. Buncic, A. Morsch, Time Projection Chamber (TPC) M. Kowalski, M. Ivanov Core Offline P. Buncic, A. Morsch, F. Rademakers, K. Safarik Muon Spectrometer (MUON) V 0 detector (VZERO) B. Cheynis A. De. Falco, G. Martinez E. Scomparin Eu Grid coordination US Grid coordination P. Cerello Transition Radiation Detector (TRD) Zero Degree Calorimeter (ZDC) L. Pinsky T 0 Detector (START) A. Maevskaya C. Blume, A. Sandoval Cosmic Ray Telescope (CRT) Photon Multiplicity Detector (PMD) Offline Board Chair F. Carminati A. Fernández B. Nandi Forward Multiplicity Detector (FMD) Web & VMC ROOT CEADEN R. Brun, F. Rademakers Detector Construction DB W. Peryt A. Maevskaya Electromagnetic Calorimeter (EMCAL) Inner Tracking System (ITS) Photon Spectrometer (PHOS) High Momentum Particle ID (HMPID) G. Odiniek, M. Horner R. Barbera, M. Masera Y. Schutz D. Di. Bari December 9, 2004 Computing Model Workshop Time of Flight (TOF) A. De. Caro, G. Valenti 45

What do we have to do by end 2005 • • • Alignment & What do we have to do by end 2005 • • • Alignment & Calibration Change of MC Integration with HLT Control of Ali. Root evolution Development of analysis environment Development of visualisation Revision of detector geometry and simulation Migration to new Grid software Physics and computing challenge 2005 Project structure & staffing Organisation of computing resources Writing of the computing TDR December 9, 2004 Computing Model Workshop 46

ALICE Physics Data Challenges Period (milestone) Fraction of the final capacity (%) 06/01 -12/01 ALICE Physics Data Challenges Period (milestone) Fraction of the final capacity (%) 06/01 -12/01 1% Physics Objective pp studies, reconstruction of TPC and ITS • First test of the complete chain from simulation to 06/02 -12/02 5% 01/04 -06/04 NEW 10% 05/05 -07/05 TBD 01/06 -06/06 20% December 9, 2004 reconstruction for the PPR • Simple analysis tools • Digits in ROOT format • • Complete chain used for trigger studies Prototype of the analysis tools Comparison with parameterised Monte. Carlo Simulated raw data • • Test of condition infrastructure and FLUKA Test of g. Lite and CASTOR Speed test of distributing data from CERN NEW Test of the final system for reconstruction and analysis Computing Model Workshop 47

CDC 04 nous sommes ici ALICE Offline Timeline CDC 05 PDC 06 Ali. Root CDC 04 nous sommes ici ALICE Offline Timeline CDC 05 PDC 06 Ali. Root ready PDC 04 Computing TDR 2004 2005 Development of new components ALICE PDC 04 Analysis PDC 04 Design of new components December 9, 2004 PDC 06 2006 PDC 06 preparation Pre-challenge ‘ 06 Ali. Root ready Final development of Ali. Root PDC 06 Computing Model Workshop First data taking preparation 48