|
SkyQuery is a prototype of a Federated database application,
implemented using a set of interoperating Web Services. This is a good
demo of how typical applications in the
Virtual Observatory will look like. This application has been developed
using Microsoft Visual Studio.NET. It won the 2nd prize on the
Microsoft
Web Services programming contest.
|
|
Astronomy is undergoing a major paradigm shift. Data in the astronomy
archives is doubling every year. More and more of the data is stored in on-line
archives. Surveys covering a large fraction of the sky are a major part of this
revolution. Our project relies on several data sets, the
Sloan Digital Sky Survey, the
the 2MASS project, and the
First radio survey.
These three archives are stored at different geographic locations.
|
|
Web Services form an integral part of this web site.
Each of the three data sets have been converted to the same framework, running
on Microsoft SQL Server, using our template from the
SkyServer, an interactive access to the SDSS data, developed in collaboration
with Jim Gray.
|
|
Spatial Matching is at the heart of our problem. We are trying to find
the same objects on the sky, detected in the different catalogs. In order to do
this, we need an efficient spatial search algorithm. Then we need to coordinate
the searches across the multiple archives, finally we need to filter the result
with the other constraints.
|
Distributed Queries are notoriously difficult. Our data sets are located
at three different places in the US, at a typical university location. We are
trying to pose the astronomy equivalent of queries, like:
Find me all the science fiction books which are carried both by Amazon
and B&N, which have similar colors, and the there is at least a $5 price difference
between the two sites.
|
|
|
