this page was started on 12/23/2003 by G.Borissov, S.Burdin,
A.Nomerotski. We update / develop this page when we have time so
if you find something that you want to know more about please don't
hesitate to contact us -
we'll be happy to help and add this information to the page.
General information
The package was written by Guennadi Borissov (U.Lancaster) and was
inspired by the software used for many years by the DELPHI detector at
LEP
for B-physics analyses. Two of us (AN & GB) worked on DELPHI and
used
the DELPHI's BANA for our theses analyses and more. The code is based
on
the D0 AATrack package used for the track reconstruction in D0reco and
therefore
has immediate access to all necessary data structures in the tracking
which
makes it very efficient. Also special care was taken to optimize the
speed
performance (ex. limited number of classes etc). We think about this
code
as physics-oriented and not necessarily object-oriented :).
Currently only tracking and muon information is available (which is
enough for most of B-physics analyses). The package also includes part
to calculate dEdx in SMT useful for particle ID. The dEdx code was
written by Sergey Burdin (Fermilab) and the dEdx information will be
available
in TMBs starting from version p15 (for now this information is
available
only for some limited datasets created from DSTs or raw data). In
general
the package was thought to be an 'open source' - style software and we
welcome people who want to contribute - we hope that simplicity of the
code makes it easier. Possible areas for contributions are the addition
of calorimeter information (jets and EM objects) and constrainted
vertexing.
The BANA package allows to do many interesting things with the Dzero
data including
easy access to any available track/particle information
define new particles from a group of children particles
define particle lists (ex. all particles, all muons, all V0's,
all J/psi's, all B => muon D0 etc.)
We use TMBs as our standard input data format. Raw data and DSTs can be
used as well but this is substantially slower. The BANA's internal data
format is called AADST. Normally we run over TMBs in SAM (could be
regular TMBs or fixed TMBs created by the CS group), select small
specialized data samples and store their AADSTs on disk at the clued0
cluster. After that the samples can be analysed with a standalone
version of BANA without the framework (even on your Linux laptop if you
install it there). As for the D0reco version we use only p14 or higher.
The following samples are currently available
Where is the Data
main thing :
AADSTs for D-skim of complete
D0 dataset
see /home/bgv/data/reco/ for
complete list of available data, it's normally
updated monthly
each subdirectory contains a dataset corresponding to some
period
of data taking (typically one week but some exceptions exist)
name of each subdirectory is YYMMDDDD_V where
YY is year (ex. 04)
MM is month (ex. 07)
DDDD is start date and end date for the dataset
(ex
2227). Dates refer to the data availability in SAM and not to the
actual
recording date.