Minutes of the Offline Meeting
June 9th, 1998

Present: T.Nakada, J.Saborido, C.Shepherd, G.Dissertori, M.Cattaneo, F.Ranjard, F.Harris, E. van Herwijnen, T.Ruf, N.Smirnov, S.Laptev, H.Dijkstra, P.Bartolini,  P.Binko, P.Colrain, A.Tsaregorodtsev, J.Harvey

Agenda:
1. Status of SICB (version 1.15) A.Tsaregorodtsev
2. ALEPH experience with BOS to Objectivity conversion        G.Dissertori
3. Discussion on "Zebra to C++" data conversion in LHCb

1. Status of SICB (version 1.15)

Andrei has presented plans to release a new version of the SICB program this week. This version is intended to be running on NT platform as well. The new release will contain several new features: new RICH pattern recognition; new vertex fit with kinematical constraints; updates in vertex trigger; interface to PDFLIB; updated particel properties table. Also some minor bug fixes and improvements will be done. Those contributors having codes ready to be released for  other are invited to commit them.

  2. ALEPH experience with BOS to Objectivity conversion

(see slides here )

Guenter Dissertori presented a work done recently within ALEPH Collaboration on moving ALEPH data from BOS/EPIO files to Objectivity database. The ALEPH data is organized in banks with ADAMO descriptions. These descriptions were used to generate C header files of classes with one-to-one correspondence to BOS tables. Each BOS bank resulted in two classes: objects (rows) and a container class. The latter was made to inherit from ooObj class of Objectivity to provide persistance. Filling
in the class objects is done with a direct memcpy operation from BOS common block.
This project was undertaken in order to gain experience with LHC++/Objectivity and eventually will provide input to ALEPH OO physics analysis program. It is in plans to have a true OO event model rather than BOS analogue in C++. Guenther pointed out that the same scheme should work equally well for the ZEBRA based banks provided that the
banks are described in some sort of a DDL.

  3. Discussion on "Zebra to C++" data conversion in LHCb

In the discussion that followed the previous presentation several points were raised. It is clear that we can relatively easily port our SICB data to C++ "ZEBRA-objects". This will
be a necessary step before filling in the event model objects still to be defined. (see also Marco's slides from the previous meeting here). The part of the event model necessary for doing physics analysis is a good candidate to be defined   and filled in first. There is obvious interest of physicists in this development. This will allow to learn such things as OO modelling, OO data storage, C++ programming, etc. A question was raised by Hans whether we can use Objectivity events in order to speed up access to them. Marco answered that this depends on the amount of data and available disk storage since the Objectivity database is only accesible from disks. It will be beneficial to define subset of data to be stored in Objectivity database to be used in the analysis. Agnieszka is one of the volonteers to look at the possible data model for the OO analysis program. Other physicist are welcome to join this activity.