Distributed Computing was the response of the High Energy Physics community to the challenge of managing unprecedently large volumes of data produced by the LHC experiments. The resulting grid computing infrastructure is now largely considered to be one of the pillars of the LHC success. The grid middleware provides common interfaces for accessing computing and storage resources. On top of it, all the LHC experiments developed their solutions for managing complex workflows of computing tasks. DIRAC is the product of the LHCb experiment which is now adopted by multiple other scientific communities for distributed computing. DIRAC services are now also provided by large grid infrastructures such as France-Grilles, GridPP or EGI, the services are available for user communities from various scientific domains. DIRAC provides solutions for both workload and data management on a large scale.
In the seminar, we will introduce distributed computing grid systems and illustrate their architecture and operations based on the example of the DIRAC Interware project. Opportunities offered by the DIRAC services to research communities will be discussed.
The seminar will be followed on Tuesday by a Tutorial session from 9 to 12 at the CPPM library:
The DIRAC Tutorial is devoted to a practical introduction into the use of DIRAC services. The participants will learn how to prepare and submit computational tasks, monitor their execution and retrieve the results. The system algorithms behind the scene will be explained. Participants will also see how to create and store new data files in a distributed computing system and make them available either locally or in remotely executing tasks. The tutorial will use a production DIRAC service which will be accessible even afterwards. The goal is to provide enough information for the participants to consider using distributed computing resources, e.g. European Grid Infrastructure (EGI) DIRAC service, for their practical research work.