Documentation for IGI Site Managers

Topics

• INFN Grid (IG) middleware release
  
• gLite middleware release
• IGI-specific middleware components (accounting - DGAS - and StoRM)
• gLite MPI configuration guidelines
• Grid operational security
• Site availability and reliability
• EGEE-III operational documentation
• Tools for site managers

INFN Grid (IG) middleware release

• Guidelines are documented here wiki

gLite middleware release

• gLite latest releases
  • gLite release 3.2
  • gLite release 3.1
• gLite supported client and server versions:
  • supported gLite client versions
  • supported gLite service versions
• gLite release documentation
• glite operational guides and FAQ

Specific IG middleware components

• StoRM is the recommended middleware for Storage Resource Management in IGI. StoRM implements the SRM v.2.2 standard and can be used for the management of both disk and tape resources. StoRM development is supported by the INFN Grid project. [Read more]
  • StoRM documentation and admin guide 
• DGAS is the Distributed Grid Accounting System of choice in IGI. DGAS provides the functionality needed for the implementation of a complete stand-alone infrastructure for computing and storage accounding of national Grids and Grid Virtual Organizations. It provides sensors for many different Local Resource Management Systems, repositories for persistent storage of usage records at site-level or nationa/regional-level, and it is comes along with a solution for a user-friendly display of accounting data, called HLRMon. DGAS and HLRmon are supported by the INFN Grid and EGEE projects. [Read more]
  • DGAS Installation and Configuration Guides
  • tour of the HLRmon portal
    
• GLUE Schema Specification, version 1.3 Final, Jan 16 2007
• GridFTP: GT 4.2.1 GridFTP System Administrator's Guide

MPI Configuration

Guidelines are documented here.

Site availability and reliability

Certified sites that are part of the production infrastructure are requested to provide a minimum monthly Availability (70%) and a minimum montly Reliability (75%). Availability are Reliability are computed montly through the Service Availability Monitoring framework. Monthy statistics are published at the following url. Sites that do not meet these minimum performance figures are requested to report on the problems encountered to the European Grid Operations Coordination bodies.

• EGEE Availability and Reliability montly reports

EGEE-III operational documentation

• Operational Procedures for ROCs and Sites
• Process for retiring gLite Grid middleware
• Process for defining supported gLite services versions
• Comprehensive collection of operational documentation.

Tools for the site manager

• a comprehensive list of operational tools for site management (broadcast tool, downtime management, SAM portal, ticketing system etc.)
• demo on the WMS Monitor tool
• LRMSinfo: tool for computation of declared installed capacity for a given site and more (number of delcared and online slots, number of online cores, type of processors installed in the farm and other information)