See this note archived at:

Dear All,

the Application Workshop discussed some of the action items from the  Madison US LHC OSG Technology Roadmap meeting. The outcome of the  discussions on metrics and goals for OSG 0.2 and the planning and  definition of OSG Release 0.4  are  provided as input to the Technical  Groups and Activities, with a goal to have them submitted for  endorsement by the council before the July consortium meeting.

The consensus at the meeting was to propose a Blueprint meeting the  29th and 30th August. The goal of the meeting will be to deliver a  Requirements  Document for Accounting, Monitoring and Information  Services for the OSG Technology Roadmap.

I thought the Applications meeting was useful and hope that the  discussions on collaboration and action items still look good after a  weekend. We had input from more than 10 VOs. People who did not attend  are invited  to read the posted talks and ask questions on the  osg-slac0605 list.

US LHC OSG Baseline Services roadmap document is
Rob and Frank will be posting a new version soon.

Define  Metrics for OSG 0.2:
    Efficiency of Grid Infrastructure and Sites (ie exclude efficiency of  applications)
    Total amount of data produced
    Total amount of data ingested by applications
    % of available CPU consumed – not a metric as impossible to measure  denominator. Sample?
    Total CPU consumed.
    Number of science papers including results from processing done on OSG
    Number of different applications run on OSG
    Ratio of number of Trouble Tickets to number of ?
    Average Time to closeout a Trouble Ticket.
    Average Time to bring up a site.
    Application Administration FTE load
    # of different users submitting jobs.
    Fraction of resource usage  that is Opportunistic

Definition of OSG 0.4 Release. Proposed for circa November 2005.  Additional services and functionality from OSG 0.2:

Edge Services Framework

    GT4 Gram as default job dispatching interface
    Policy for quota’d space per VO (50G each)
    Support for Late binding WMS

Workload Management System
    Support for Late binding
    Job inspection service
    Guaranteed job exit codes – batch job executor sends exit code through  xxx

Service Catalog
    Extensible information about service and resource
    Dynamic registration

    SE based on SRM
    Storage role based access control

    resource information repository, interoperable with LCG and partner  Grids and with VO specific monitoring infrastructure.
    Metrics instrumentation and presentation


Federation infrastructure with partner grids.


VO Management
    FQAN name space policy
    Redundancy and fault tolerance (e.g. for GUMS and VOMS implementations)
    Support for multiple implementations (ie not just INFN VOMS)