The Collaborative Simulation MetaData Management-CoSiMM
We present here our ideas for developing and evaluating key elements of a Virtual Research Environments supporting scientific collaborations, based on experiences in a pilot project with the CLUES collaboration.
The CLUES (Constrained Local Universe Simulations) project is an international collaboration of more than 30 scientists with the goal of reproducing the local universe with constrained cosmological simulations. Due to the computational power and time needed, the simulations are performed at various supercomputing centers worldwide. This complicates the data exchange between many scientists.
One of the key elements of the Virtual Research Environment is providing computational resources dedicated to assist the hosting, exchange and post-processing of simulation raw data. In order to manage the distributed data we designed a simulation metadata & storage database. It keeps track of data creation, location, access and post processing. It stores the relation of data sets. The user friendly interface allows registered users to enter new data sets and manage them in a simple way.
The programming model is based on a combination of PHP, AJAX and MySQL, allowing to offload the server from unnecessary requests with AJAX, making the communication with the server asynchronous. The metadata handling and data management assists the users to focus on their scientific targets and provides new ways to query and compare the post processing results.
Return to poster list