This dataset is a subset of STAR TAG data produced from year 2000.  This
data set was used in some tests of the FastBit bitmap index system, in
particular, the performance of the compressed bitmap schemes.

The dataset can be thought as a table, where each column is written to a
separate file.  There are 12 columns in this dataset, they are extracted
from a larger dataset containing about 500 columns.  The 500 columns are
known as tags, and are processed from a raw dataset of much larger size.
More description about the dats is available from STAR website
<http://www.star.bnl.gov>.

If you use this dataset, please refer to
<http://www.star.bnl.gov/public/central/collaboration/acknowledgments.php>
for requsite acknowledgement.

*NOTE* The binary data in star2000.tgz.bz2 is in big-endian.  It was
 produced on a Sun SPARC workstation and is *NOT* usable on x86 based
 machines without transformation.

Variable names and type in star2000.csv (table.tdc in star2000.tgz.bz2
contains a longer names used by the physicists in the description fields)
charge		float
clus		int
dst		int
hist		int
mEventNumber	int
mEventTime	double
mRunNumber	int
NLb		int
qxb		float
tracks		int
vertex		float
zdc		int

To use ardea.cpp to convert star2000.csv, use the following string as
metadata

PATH-to-ARDEA -m "charge:f, clus:i, dst:i, hist:i, enumber:i, etime:d, rnumber:i, nlb:i, qxb:f, tracks:i, vertex:f, zdc:i"