ATLAS PPDG quarterly report Q2'01 (Apr-Jun) =========================================== Distributed data management --------------------------- [ATLAS distributed data manager is now to be an 'activity' as I understand it.] The principal PPDG year 1 deliverable for ATLAS in the first year is the delivery of a production distributed data service deployed to users. The distributed data manager activity in ATLAS PPDG is developing the system which will be deployed to meet this deliverable. The production distributed data service is to exist between CERN, the ATLAS Tier 1 facility at BNL, and a number of US ATLAS grid testbed sites (some or all of ANL, LBNL, Boston University, Indiana University, U Michigan, U Oklahoma, and UT Arlington). The objective is a multi-point U.S. Grid (in addition to the CERN link) providing distributed data services as early as possible. Development of the distributed data manager was initiated during the period, based on the DBYA prototype introduced in the last report as a starting point. DBYA provides ATLAS-specific infrastructure and metadata, and the infrastructure to vertically integrate the data manager from the grid toolkit components employed through to the ATLAS-specific interfaces by which the system is used. Grid toolkit components will be progressively integrated into the system. During the period a facility to load the file replica metadata of the system into the Globus replica catalog was developed. Over the next few months we expect to extend the functionality of the system from the present passive cataloging to active replication of files between grid sites using GridFTP. We provided input to the Replication Requirements document in development, based in part on DBYA prototyping experience. We are continuing to use ATLAS tile calorimeter test beam raw data in an Objectivity database to test the data transport and data replication tool sets. A Globus replica catalog running at ANL with catalog query and update clients running at BNL was successfully demonstrated and exercised, both from a command line interface and from C programs. This is a first step in a program to demonstrate grid-enabled data access from Athena, the ATLAS experiment's control framework, in the coming quarter. An abstract for a paper describing this work was accepted for presentation and publication in the proceedings of CHEP'01, Beijing, September 2001. See http://atlassw1.phy.bnl.gov/dbya/info for more information on the distributed data manager prototype. US ATLAS Grid Testbed --------------------- The US ATLAS Grid Testbed is now composed of 8 US ATLAS sites, ANL, BNL, LBNL, Boston U, Michigan U., Indiana U, Oklahoma U., and U of Texas at Arlington. Each provides a dedicated Globus 1.1.3/4 gatekeeper and additional computational resources for the testing and evaluation of GRID technologies as applied to ATLAS. During the period the Globus services have been in stable operation and the focus has been on the installation of ATLAS software and infrastructure at the testbed sites. During the summer and fall the focus will shift to applications testing. See http://www.usatlas.bnl.gov/computing/grid/ for more testbed information. Monitoring ---------- A simple tool kit based on iperf was created to facilitate performance monitoring and tuning among the different testbed sites. A low bandwidth problem between BNL and NERSC was diagnosed and a solution is in development. A web site gathering network performance information is in development at http://yu.rhic.bnl.gov/~cricket/cricket/grapher.cgi Distributed job management -------------------------- A secondary year 1 objective for ATLAS, of growing importance in the out years, is distributed job management. During the period an ATLAS GriPhyN effort was initiated (Saul Youssef, BU) to study and test the capabilities of Condor to manage a hierarchical job management infrastructure incorporating the various tiers of grid sites. The ATLAS PPDG team is in contact with this effort, and discussions are underway on a collaborative effort with PPDG. See http://physics.bu.edu/~youssef/atlas/notes/ for more information on the Condor scheme being investigated. Manpower -------- Most of the ATLAS effort is to come from new hires or internal transfers, supported by relatively small fractions of existing personnel. In the BNL physics applications software group (1.5 PPDG FTEs) we have hired two new software developers who will start in September. Part of their time will be spent on PPDG. We have opened a position for a further hire which will complete the team. In the BNL Tier 1 (.5 PPDG FTEs) a very well qualified recent hire, Dantong Yu, is fully up to speed. ANL (0.8 ATLAS PPDG FTE) plans to acquire an existing person from within the lab, as yet unidentified. During the period ATLAS and PPDG agreed on an ATLAS-CS liaison person. Jennifer Schopf, who has recently joined Ian Foster's group at ANL, will fill this role. Jennifer is also the ATLAS-GriPhyN liaison which will provide an important linkage between the ATLAS programs of these two projects.