Data model for SDSS-III

General description

Here we describe the SDSS-III data model, which is archived in the datamodel product at "$DATAMODEL_DIR/datamodel/files". One can access the data in two ways:

• Data model directory structure
• Data model file index

Glossary of Common Terms in the SDSS-III Data Model

Data model directory structure format

In the data model path, there are files of the form:

 [ENVVAR]/[..path...]/[filename].html

The ENVVAR represents the environmental variable name of the root directory for each data type. Each .html file represents the generic name of a type of file (plus .html), as well as its position in the overall directory structure. The paths contain generic names for the type of subdirectories, which should come from the following list:

TODO: expand this list to include new names.

 RUN
 RERUN
 CAMCOL
 MJD
 PLATEID
 PLATERUN
 DESIGNID

If an integer is appended (as in RUN6) it implies that the name will be an integer, and should be expressed in that number of digits (filled out with leading 0s).

If an XX is appended, that means that the last two digits of the number are replaced by XX (so we can group files and subdirectories in batches of 100).

There is a template.html file in this directory that gives an example of how to write a description of a single file, and the sort of information that can/should be included. Note that the template is valid XHTML 1.1, & this standard will be enforced on all html files in the data model. The W3C Validator will allow you to validate html before checking it in.

There are some special classes of parameters and flags that are repeated several times, and have a common meaning throughout the survey. The glossary file explains them. Items in the glossary should be referred to with URLs of the form:

    href="/datamodel/glossary.html#asinh"

---