The CosmoSim database provides data from cosmological simulations in databases which can be queried directly using the Structured Query Language SQL. Since the amount of data which such simulations produce nowadays exceeds the Terabyte range, the full data set is too large to be kept as a local copy for each user. Instead, the user can retrieve only the data subset he or she is interested in. Large observational surveys like the SDSS have already made the first steps in this direction. Having the data directly available via SQL proved to be a very useful concept. With the Millennium Run Database developed within GAVO (German Astrophysical Virtual Observatory), this concept was extended to simulation data as well, allowing to exploit the data much more effectively. In return, this gave rise to many fruitful collaborations and hundreds of articles.

Here, we are providing data from large scale simulations which are in extend and particle numbers similar to the Millennium Simulation, but with different volumes and resolution. Each simulation is describen in the Database tables section, for example BigMDPL simulation. In this view, you can find the tables available for any simulation. For getting full access to these data via the Query form, please register first.

A detailed column description for each type of database table is given in the next chapter of the documentation: Data structure. The Technical documentaion will provide you with examples and tutorials to get you started.

Be aware, that some things have changed with this new Django Daiquiri and the Postgres technology behind it all. See CosmoSim - migrated for an overview on the differences.


Please fill out the registration form to register for a CosmoSim database account. This will provide you with following features:

  • Access to all simulations published in the database
  • Long queue for queries is enabled, with less strong timeout restriction
  • Your own private user database cosmosim_user_<username>, where all your query results are stored

It is also possible to submit queries as a guest. You can use the the short queue (timeout limit: 30 seconds) to try out some examples. The results will then be stored in the public-guest /anonymous database, where any other guest can view and remove them. The results are removed from the anonymous database regulary.

When you fill out the registration form, we also ask for some information about you, especially about your affiliation and your intended use (“description” field). This information is mainly used for statistical purposes as well as to get an impression of what users would like to achieve with the database. This helps us to get to know in which direction we should develop our services further.