HEP Applications. At the beginning of the project, and throughout its lifetime, there will be a continuing dialogue between HEP applications and middleware developers for the processes of requirements gathering , grid service design and the practical evaluation of software for its functionality and performance. Applications will also be part of the testing and acceptance processes for the middleware. We have learned from DataGrid that all of these processes need to be ongoing to avoid ‘big bang’ effects due to infrequent overall system status reviews. This ongoing dialogue will be accomplished by HEP applications participation in the architecture, design and system testing groups. To facilitate this dialogue the HEP applications staff will have to work as a coherent team, whilst still maintaining experiment specific support, but looking for commonality in the support and associated tools.
HEP Applications. The 4 LHC experiments (Atlas, XXXXX, CMS, LHCb) each have developed functioning distributed computing systems over the past few years. In the past 2 years they have made good progress in integrating Grid technology into these systems, notably in Europe with DataGrid, DataTAG and Nordugrid software, and in the US with software from the various US projects. The experiments have also produced significant grid software, for example the Alien system of XXXXX and the DIRAC system of LHCb. They will continue this work in the context of the LCG project which aims at providing a production service to the experiments. The main experiment applications are currently the generation of vast quantities of simulated data necessary for the preparation of the experiments and LHC start-up in 2007, and their reconstruction and analysis. In addition to this highly organised data production work there is the computing associated with thousands of physicists and engineers all over the world accessing the data for individual analyses. The load characteristics associated with the analysis work are highly variable and will place very special demands on the grid for resource allocation and quality of service. The principal function of the team of 8 people (see ‘funded/unfunded participants’ for group organisation), employed by and based at CERN, will be to work with the 4 LHC experiments, to ensure their requirements are taken into account, to give practical help in interfacing the experiment application to grid services, and to evaluate the performance of the software deployed within the LCG service environment, as well as in pre- production testbeds. It is foreseen that non-LHC experiments, as they have done in DataGrid, will participate in EGEE. Current examples are US based experiments, BaBar and D0, which are currently taking production data and are basing their medium-term computing strategy on distributed computing resources served by a powerful Grid. These will be supported on a best-efforts basis.
HEP Applications. CERN (funded and unfunded) The 4 posts funded by EGEE will be placed with the CERN EP/SFT group. This group has representatives from all the LHC experiments, including the software architects, and also includes coordinators for several key LCG activities, including the applications area. CERN will provide 4 FTEs in addition to the 4 FTEs funded by XXXX, giving a team of 8. There are several people already within the LCG project, funded by CERN, working in this area. X. Xxxxxx, the DataGrid HEP Applications WP Manager, will form part of this effort for the start-up phase of the EGEE project. Strong working links with the 4 LHC experiments will be established, and two people will be assigned to each experiment. The team will be managed as a coherent whole allowing some flexibility in support of the experiments and the establishment and execution of common goals. Members of the team will work with the experiments to help define and execute the work programmes associated with integrating the experiment systems with EGEE middleware. They will evaluate the performances obtained on the LCG infrastructure and provide material for the evaluation reports. Members of the team will contribute to the work on the common application layer specification, liaising closely with the LCG ARDA and GAG activities It is planned that 80% of the effort of team members will go towards experiment specific activities, and 20% for common activities such as architecture, design, and the production of the EGEE deliverables. Russian Institutes (funded and unfunded) The planned activities of the participating Russian institutes are as follows: ▪ IHEP - Management of ATLAS VOs in Russia; Support of XXXXX, CMS and LHCb pilot applications; Identification of early users; ▪ ITEP - Management of LHCb and XXXXX XXx in Russia, Support of ATLAS and CMS pilot applications; Identify early users; Introduction of new user communities; ▪ JINR - Support ATLAS, XXXXX, CMS and LHCb pilot applications; Identify early users; Introduction of new user communities; ▪ PNPI - Support HEP pilot applications (ATLAS, CMS, XXXXX, LHCb) for St-Petersburg region, identification of early users; ▪ RRC KI - Support of XXXXX pilot applications. Lead partner for Nuclear Fusion GRID applications in Russia. Identification of early users. Introduction of new user communities; ▪ SINP-MSU - Management of CMS VO in Russia; Support of ATLAS, XXXXX and LHCb pilot applications; Introduction of new user communities;
