| Bart Scheers | 0818c337d503 |
7 years ago
|
|||
| Bart Scheers | 72890999e24d |
7 years ago
|
|
Running LSST baseline queries in MonetDB
See blog post LSST Queries in MonetDB for further details.
You need to have several LSST csv files (with artificial data), which can be requested from the LSST DM team.
MonetDB
Run the installation script <tt class="docutils literal">install_monetdb.sh</tt>, start the daemon and create a database instance.
Loading
The python load script <tt class="docutils literal">load_lsst_data.py</tt> defines the database schema with table and partition defintions and loads the csv files into the database.
Running queries
Two shell scripts run the queries in a hot and cold mode <tt class="docutils literal">hotloop.queries.sh</tt> and <tt class="docutils literal">coldloop.queries.sh</tt>, resp. The averages and standard deviations of ten runs are reported in the log files
Plotting results
Use the python script <tt class="docutils literal">plot_baselines.py</tt> to plot a bar chart of the results in comparison with the S15 MySQL results.
More info
For more info send an e-mail to bscheers at cwi.nl.