Deployment Team Deployment – Central Management Team Takes care of the deployment of the release, certificates the sites and manages the grid services • Detailed deployment plan • Installation notification via web • Use of trouble ticket system to notify installation problems • Cvs repository of site configuration • Separate grid services for certification phase • Middleware and application specific test suite Site/resource map of INFN-Grid CMS T.2 T2/3 Atlas T.2 T2/3 Alice T.2 T2/3 LHCb T.2 T2/3 Babar VIRGO grid.it resources general purpose resources TRENTO MILANO UDINE PADOVA TORINO LNL PAVIA TRIESTE FERRARA PARMA GENOVA BOLOGNA CNAF PISA FIRENZE S.Piero PERUGIA LNGS ROMA2 ROMA L’AQUILA LNF SASSARI NAPOLI BARI SALERNO CAGLIARI COSENZA PALERMO CATANIA LNS LECCE LCG participation: -Tier1 cnaf, -Tier2 in MI, TO, LNL, Roma1, NA Deployment plan for INFN-GRID 2.0.0 (*) • Date: March 1st 2004 – Goals: • Upgrading the INFN production grid to INFN-GRID 2.0.0 Make available for all VOs a grid for production activities – Upgrading type: • Invasive – List of changes: • New WorkLoad rpms more stable. New DataManagement and new RLS compatible with POOL format – Sites involved: • Bologna-CNAF Bologna Padova Legnaro-Tier2 Torino-Tier2 Milano-Tier2 Pisa Catania Cagliari Genova Bari Napoli Trieste Ferrara Roma1-Tier2 Roma1-VIRGO Lecce INAF-Catania Genova Perugia Pavia INAF-Trieste – Available grid services: • ResourceBroker/BDII - CNAF ResourceBroker/BDII - Padova TOPMDS - CNAF RLS - CNAF RLS for CMS - CERN – Note: • All Tier2 sites (Torino, Milano, Legnaro and Roma1) should temporarily register in the GRID-IT TOPMDS (edt001.cnaf.infn.it). The BDII will also be populated by the LCGSouth TOPMDS in order to make available Tier1 site to the whole INFN production grid. The RLS for CMS will be managed by LCG organization • Installation • • • • • • • • • Please follow these steps to install a site with the latest distribution: Step 1: Requirements Step 2: Install your LCFGng server Step 3: Install your site Once everything is successfully installed, you can jump to the Testing section. • » Manage your site Contact Central Management Team to join the Grid • Once completed the installation of the upgrade of your site, you have to inform the Central Management Team completing this form. Please remember to commit and tag in CVS your site configuration (instructions are provided in installation and upgrade guides. • Please fill in all the following fields to complete the procedure to join the GRID. • Fields with (*) are mandatory. (*) Your site name (Institute-City' - e.g. INFN-Bologna): (*) ComputingElement hostname (FQDN): (*) JobManager: (*) Queues (Normally: short,long,infinite): (*) StorageElement hostname (FQDN): (*) WorkerNodes hostname (FQDN) list: (*)Version: (*) Operation type: Installation completed Upgrade completed GrAT Grid Application Test (GrAT) origins: • GrAT is originated by the exigency to enable new tests for the emerging and existing applications of each Virtual Organisation • GrAT is designed to aid site-managers and GOC to test installed grid applications. GrAT working • Tests are centrally submitted to Italian sites using a web interface. Terminal Browsing UI • Results are stored in a MySQL Database MySQL Database Server WEB • Results can be consulted on line. CE SE Web Interface 1/3 A WEB portal gives an overview on Italian testbed. A summary table contains last test results for each site grouped by Virtual Organisation. Web Interface 2/3 Job details are available on line: Start and finish time Log and output test Worker Node hostname • Grid Service Infrastructure snapshot The Grid Service Infrastructure is distributed as much as possible in different sites. In this way: – More back-up solutions – More Fault Tolerance – No bottlenecks • Each site which hosts a grid service must guarantee: – Dedicated and compatible hardware with software requirements. – Support for any kind of trouble in the declared response time • Operations: – The local site administrator install and configure the grid service according to release notes – The Grid Service Management Team certifies the grid service • The Grid Service Management Team can provide grid services ad hoc – BDII with a particular set of sites – Resource Broker for particular activities • Failure discovering – GridICE – Ticket System Grid Service Infrastructure Grid Service Infrastructure Monitoring • Provided by GridICE • Monitoring of: – Host – Daemons • One tool for distributed grid services • All grid services must be monitored