Evolution of the Atlas Data and Computing Model for a Tier2 in the EGI Infrastructure
Fernández Casaní ÁLVARO, Miguel Villaplana Pérez, S. González de la Hoz, José Salt Cairols · 2012
On behalf of the ATLAS CollaborationSince the start of the LHC pp collisions in 2010, the ATLAS computing model has moved from a strict hierarchical design, where every Tier-2 had a liaison and a network dependence on a Tier-1, to a more meshed approach with direct connections between Tiers.Evolution of ATLAS data models requires changes in ATLAS Tier-2 policy for the data replication, dynamic data caching and remote data access.It also requires rethinking the network infrastructure to enable any Tier-2 and associated Tier-3 to easily connect to any Tier-1 or Tier-2.The Tier-2 disk space is used for real, simulated, calibration and alignment, group, and user data.A cache disk space is needed for input and output data for simulations and production jobs.A number of concepts and challenges are raised in these proposals, and in this contribution we show how these changes affect an ATLAS Tier-2 and its co-located Tier-3 as a part of the EGI infrastructure.We will present the Tier-2 and Tier-3 facility setup, the data distribution and the arrangements proposed to fulfil the requirements coming from the new model, especially the requirement for any site to be the source for data replication.A given site can receive datasets from any other site "on demand", based on usage patterns, and possibly using a dynamic placement of datasets managed centrally and unused data can be centrally removed as well.The data access for users either with grid or local tools will be presented, using the EGI infrastructure and procedures, and the middleware glite flavour that is being provided by EMI releases.We use an example of a real physics analysis to show how users are working, to check the readiness of the tools and its performance within the changes being adopted coming from the evolution of the model.