The ganga user interface for physics analysis and distributed resources

A physicist analysing data from the LHC experiments will have to deal with data and computing resources that are distributed across multiple locations and have different access methods. Ganga helps by providing a uniform high-level interface to the different low-level solutions for the required tasks, ranging from the specification of input data to the retrieval and post-processing of the output. For LHCb and ATLAS the goal is to assist in running jobs based on the Gaudi/Athena C++ framework. Ganga is written in python and presents the user with a single GUI rather than a set of different applications. It uses pluggable modules to interact with external tools for operations such as querying metadata catalogues, job configuration and job submission. At start-up, the user is presented with a list of templates for common analysis tasks, and information about ongoing tasks is stored from one invocation to the next. Ganga can also be used through a command line interface. This closely mirrors the functionality of the GUI, allowing easy transition from one to the other. This paper describes the Ganga design and functionality, and illustrates its use in the distributed analysis systems of the LHCb and ATLAS experiments in the context of their 2004 data challenges.