Minutes of the PandaRoot meeting, 11th of November, 2009, EVO, 10:00-12:30

Participants

• Mohammad Al-Turany
• Ola Biegun
• Gianluigi Boca
• Susanna Costanza
• Dipanwita Dutta
• Klaus Goetzen
• Christian Hoeppner
• Radek Karabowicz
• Ralf Kliemt
• Bertram Kopf
• Soeren Lange
• Dima Melnychuk
• Vanni Rajan Suyam Jothi
• Pavia Group
• Tobias Stockmanns
• Stefano Spataro
• ...

General

• Next Collaboration Meeting: Besides the general computing session taking place on Thursday, there will be an additional meeting on Wednesday which is dedicated to the tracking software. Ralf Kliemt initiates this meeting and the idea is to discuss/brainstorm about a concrete implementation of track merging.
• Disc space at GSI: Soeren pointed out that a lot of disc space at GSI is allocated. Most of the space is occupied by the PandaRoot data challenges as well as by the Physics Book data. While the responsible persons of the data challenge will take care of deleting the PandaRoot data, Bertram will contact the Physics Book production manager Marc in order to discuss the possibility to copy the Physics Book data to the Bochum event store.
• CAD converter for the EMC geometry: Stefano mentioned that a CAD converter is available to convert specific shapes from the IGES format - this is obviously a special STEP format of a CAD file - to the PandaRoot detector description. This converter was originally written by George Serbanut and supports especially the EMC crystal shapes. During one of the next EMC subgroup meetings it will be discussed if this converter can be used for the latest drawings from Philippe Rosier.

Proxy Dictionary

Bertram proposed to use a proxy dictionary in PandaRoot. He tried to summarize the idea and the features of a proxy dictionary. The most important point is that it is able to treat all kinds of objects in a type-save way. Nasty casts in the user code are not needed anymore what allows OO programming and the recognition of bugs at compile-time. The dictionary can be used for example to share objects with event-live-time or references to them between different tasks. Another use case would be to introduce a proxy dictionary for job-live-time objects as a kind of API to the conditions database. There was an extensive discussion about the new code and it was not clear for most of the participants what the benefits in comparison to the existing structure are. Everything could be done with Root I/O. Even complicated associations between objects can be realized in a non-OO way by storing indexes instead of treating objects. The existing runtime database can just handle parameters. For most of the participants this seems to be sufficient and would fulfill the Panda software requirements. Therefore it was agreed upon to postpone the decision of using the proxy dictionary and to deliver more pros and cons.

Activities of the Users

There are a couple of users who are currently doing some physics performance and detector studies. Marius is studying the radiation dose for the MVD, Susanna takes care of detector studies for the STT, and the Torino group with George is analyzing the Dell-Yan process with the intention to optimize the Muon detector. In addition Martin from Giessen started with the analysis of X(3872). The idea is to focus the upcoming meetings also on the users activities.

Next EVO Meeting

The next PandaRoot EVO meeting will take place on the 25th of November at 10h00.