At the ASTRON Hackathon

Statistical querying of astronomical databases

We are working on a dataset of 2M (unique) sources from the IPHAS survey. The csv file is available from here.

Install & run

• DIY: Install the MonetDB database (get it from here)
• DIY: Start the daemon (monetdbd)
• Create and start the database instance (createstartdb.sh). The database instance is named hackadb. Note that we set embedpy=true.
• Create the runcat table into which the data from the csv file will be loaded, and actually load the data (load.sh).

Now all data should be loaded

Create the UDFs

• Run createudfs.sh to create the sql/python udfs. One calculates the chi squared probability and the other renders the healpix number given the map resolution that contains the point position on the sphere.
• See it in action, runhealpyudf.sh that executes the example query (sql/examplehealpix.sql) using the healpy functionality.
• The python script plotskycov.py creates a sky-coverage image of the sample of sources