Transcript griddemo
Edinburgh University Birmingham University The BaBar experiment is expanding our understanding of antimatter. Running at an accelerator in Stanford, it gathers data on the decay of the rare B particles. Bristol University We have now recorded over 100 million B decays. Each provides a lot of data (20 Mb+). This data is studied by a team of over 500 physicists from many countries, including a strong contingent Brunel University Imperial College from the UK Liverpool University Data is distributed across many sites, and may be Manchester duplicated. Processing must University be spread across available resources: we use dedicated Royal CPU farms at 9 UK institutes, Holloway and also Lyon, Karlsruhe, and Bologna as well as Stanford. Rutherford Grid technology is used to Laboratory keep track of what data is stored where, and what Queen Mary computing power is best College placed to analyse it. It provides the power and flexibility we need. For more information see http://www.gridpp.ac.uk/babar/jobmake/grid..html The BaBar sites are monitored. A live map shows which are running and which are down. It shows the number of machines in use at each site, and the progress of event simulation jobs. Monitoring Data location + Job submission + Job retrieval ? A physicist selects events of a particular type that they are interested in, and uses a web browser to ask which sites have files of this data. Sites are given in order of preference: they will probably want data from their local site, and go to remote sites only for files that don’t exist locally. The system handles this prioritisation. Jobs are launched to analyse this data at local and remote sites. All authorisation and authentication is done through the Grid using a web browser so all resources are available without further hassle. Output from the jobs is collected and merged back together to give a complete and seamless set of results. As Grid technology progresses over the next few years,this will become smoother and easier to use, as the system will take over responsibility for the decisions that at preset have to be made by the physicist.