Scientific Information Retrieval System Using Storage Resource Broker

Abstract

The Blue Brain Project is aimed at achieving a biologically accurate simulation of the cortical column. It is collaboration between Swiss Federal Institute of Technology, Lausanne and IBM and is a prominent example of a neuroscience project that extensively draws on computer science. A large amount of data related to electrophysiology, morphology and genetic properties of neocortical neurons is already present in project’s data bank. In future more data is expected to be acquired from several places around the world. The data is in the form of files and metadata is associated with the files. The Blue Brain Project (BBP) Database uses Storage Resource Broker (SRB), a distributed file system integrated with a relational database, as its data repository. SRB provides efficient storage of the large number of files as well as powerful and fast querying performance. Software tools have already been developed for data storage, manipulation and analysis. In order to integrate these tools, a software development framework has been proposed and developed, based on component based software engineering. A cross-lab data access and integration system has been provided by using the state of the art EMCAT and Zone Federation concepts of SRB. Furthermore, a keyword based searching system has been built on top of the cross-lab data accessing system. This searching system has been empowered with an ontology support. The proposed system is generic and can have applications beyond the BBP.