MEG Software Status Framework Large Prototype software updates

  • Slides: 19
Download presentation
MEG Software Status üFramework üLarge Prototype software updates üDatabase üROME üMonte Carlo Simulation MEG

MEG Software Status üFramework üLarge Prototype software updates üDatabase üROME üMonte Carlo Simulation MEG Software Group Shuei YAMADA @ MEG review meeting, 2 July 2004 1

Framework Online n ROME - Root based Object oriented Midas Extension Offline / reconstruction

Framework Online n ROME - Root based Object oriented Midas Extension Offline / reconstruction ROOT/C++ based See Corrado’s presentation n Monte Carlo simulation n GEANT 3 / FORTRAN 77 Shuei YAMADA @ MEG review meeting, 2 July 2004 2

New DAQ for Large Prototype DAQ computer divided into 3 PCs: 1. Front-end for

New DAQ for Large Prototype DAQ computer divided into 3 PCs: 1. Front-end for trigger & data logging 2. Back-end for online analyzer, ROOT server 3. Run control, slow control & event display Shuei YAMADA @ MEG review meeting, 2 July 2004 3

Software Modifications for Large Prototype Online n n ROOT based analyzer Monitor online histograms

Software Modifications for Large Prototype Online n n ROOT based analyzer Monitor online histograms thru network Offline n n n Two analyzers: ROOT based & ntuple based kumacs are merged in the ROOT-analyzer Database : run catalog & connection table Testing to use ROME Shuei YAMADA @ MEG review meeting, 2 July 2004 4

Another CEX beam test at p. E 5 Further modification required n n DAQ

Another CEX beam test at p. E 5 Further modification required n n DAQ using wave-form digitizer Database update during data taking w Run catalog w Hardware configurations Ø Cable connection, HV value, … Shuei YAMADA @ MEG review meeting, 2 July 2004 5

RDBMS Relational Database Management System w Table format database consists from several tables w

RDBMS Relational Database Management System w Table format database consists from several tables w Tables are connected with relations CD Tune title tune_title artist_name company_name title composer_name rating Artist Composer name instrument name category Shuei YAMADA @ MEG review meeting, 2 July 2004 6

SQL Standard Query Language n Standard format of statement to access to database Ø

SQL Standard Query Language n Standard format of statement to access to database Ø SELECT title FROM Tune WHERE composer = “Juno Reactor”; Ø INSERT INTO Tune(title, rating) VALUE (“High Energy Protons”, 3); Ø UPDATE Tune SET rating=5 WHERE title = “High Energy Protons”; Ø DELETE Artist WHERE name = “DJ 19”; Shuei YAMADA @ MEG review meeting, 2 July 2004 7

My. SQL Network access Open source Large : 10000 ch x 8 M runs

My. SQL Network access Open source Large : 10000 ch x 8 M runs Fast : <10 sec for 256 ch x 1000 run pedestals Multi platform : Linux, Macintosh, Solaris, Windows… Server replication Italy PSI Russi a Japan XML database n if no network connection available Shuei YAMADA @ MEG review meeting, 2 July 2004 8

How to access? UNIX command C/C++ n DAQ ROOT n Analysis tool Web browser

How to access? UNIX command C/C++ n DAQ ROOT n Analysis tool Web browser n PHP/Perl/Ruby ODBC/JDBC n MS Excel, Access Shuei YAMADA @ MEG review meeting, 2 July 2004 9

Large Prototype database Connection table Run catalog Shuei YAMADA @ MEG review meeting, 2

Large Prototype database Connection table Run catalog Shuei YAMADA @ MEG review meeting, 2 July 2004 10

ROME Root based Object oriented Midas Extension • Tool for Event based Data Analysis

ROME Root based Object oriented Midas Extension • Tool for Event based Data Analysis • Fully Object Oriented • Root based • Full connection to the Midas Environment • Online and Offline • Based on Tasks, Containers and Folders for a good Data and Program Structure • Experiment independent Base Classes • Experiment dependent Classes are generated out of simple XML-Files • The Users write only experiment specific code (physics) • Administrative code is implemented in the generated code • Self Documenting Code • Self Linking Project Shuei YAMADA @ MEG review meeting, 2 July 2004 11

ROME Objects Folders Tasks • Objects, where data is stored in • Tasks are

ROME Objects Folders Tasks • Objects, where data is stored in • Tasks are objects, which provides actions • Stores the data of one detector (or sub detector) component • They make calculations • Hierarchically arranged • Data inside of Folders is structured • Store and read data in folders • Fill trees and histograms • Hierarchically arranged • Task also own histograms Trees • Data Objects : only written, never read Histograms • Used to write data on files • Graphical Data Objects : only written Steering Parameters • Belong to one Task • Task steering • Framework steering Shuei YAMADA @ MEG review meeting, 2 July 2004 12

Interconnections Disk Read (any Format) (Input) Read Histograms Fill Tasks Fill Histograms Folders Fill

Interconnections Disk Read (any Format) (Input) Read Histograms Fill Tasks Fill Histograms Folders Fill Histograms Flag Read Trees Write (ROOT) Disk (Output) Shuei YAMADA @ MEG review meeting, 2 July 2004 13

The ROME Environment • experiment independent ‘ROME’-Classes o Base classes for the generated, experiment

The ROME Environment • experiment independent ‘ROME’-Classes o Base classes for the generated, experiment dependent classes • ROMEBuilder o Builds all experiment dependent classes out of simple XML Files o XML files describe Tasks, Folders, Trees, Histos, Steering Parameters and Midas Banks o Links the generated project o Documents the generated project ROME Environment ROMEclasses ROMEBuilder Project ROME classes Exp. classes Executable Documentation XML File Shuei YAMADA @ MEG review meeting, 2 July 2004 14

Tasks XML File <Experiment Shortcut> <Task> <Sub. Task Name=“Task Name”> <Histos> <Histo Name Type=“Variable

Tasks XML File <Experiment Shortcut> <Task> <Sub. Task Name=“Task Name”> <Histos> <Histo Name Type=“Variable Type” Array. Size=“Array Size” Title=“Title” Folder. Name=”Histo Folder Name” Folder. Title=”Histo Folder Title” Number. Of. Bins. X=“# of X bins” XMin=“X min” XMax=“X max”/> </Histos> </Sub. Task> </Experiment Shortcut> Code void void [Experiment [Experiment Shortcut]T[Task Shortcut]T[Task Name]: : Init() { } Name]: : Begin. Of. Run() { } Name]: : Event() { } Name]: : End. Of. Run() { } Name]: : Terminate() { } Fill[Histo Name](double value, double weight) Draw[Histo Name]() Get[Histo Name]Handle() Single Histograms Fill[Histo Name]At(int index, double value, double weight) Draw[Histo Name]At() Get[Histo Name]Handle. At(int index) Shuei YAMADA @ MEG review meeting, 2 July 2004 Histogram Arrays 15

Benefit of a Generated Framework • Consistent Program Structure o All classes look the

Benefit of a Generated Framework • Consistent Program Structure o All classes look the same o Better readability • Less Handwritten Code of a class is written once (in the builder) and reproduced many times • Easier Maintenance o Modification are done once (in the builder) and then available in the whole framework C++/OO knowledge not required for programmers, just write code in C Shuei YAMADA @ MEG review meeting, 2 July 2004 16

Monte Carlo Simulation GEANT 3 / FORTRAN 77 Requires n n n Sync LXe

Monte Carlo Simulation GEANT 3 / FORTRAN 77 Requires n n n Sync LXe geometry with PMT folder design Geometry Database MC run catalog Database w m eg signal MC w Background MC n n Digitization Trigger simulation w How to combine signal & BG into one “triggered”-event? n n Programmers’ manual and End-users’ manual Human resource assignment Shuei YAMADA @ MEG review meeting, 2 July 2004 17

MC schedule/man power July By end of April 2004 n Liq. Xe geometry: S.

MC schedule/man power July By end of April 2004 n Liq. Xe geometry: S. Yamada/F. Cei September By end of June 2004 n Digitization XE: S. Yamada/F. Cei DC: H. Nishiguchi TC: P. Cattaneo Shuei YAMADA @ MEG review meeting, 2 July 2004 18

Summary Offline framework n use ROOT/C++ Another CEX beam test using new software n

Summary Offline framework n use ROOT/C++ Another CEX beam test using new software n n n ROME - ROOT/C++ based online software My. SQL database Wave-form digitizer Monte Carlo Simulation n GEANT 3 / FORTRAN 77 Shuei YAMADA @ MEG review meeting, 2 July 2004 19