User Stories for the Computing Element

• Story 1
  • Card: A user can search for the number of used and total computing resources for a certain computing element
  • Actor: site administrator, grid operator center
  • Conversation:
    • keep in mind the end-user viewpoint (job slot) and the physical resources; job slots should have a different user story
    • what is the unit of computing resource that can be free/used/unavailable
    • when a unit of computing resource is free/used/unavailable
    • what do you count in the total that is what draws the border of the system; how do you group resource into a computing element
  • Acceptance Test:
    • we need acceptance test to check for double-counting
    • we need acceptance test to check if multi-core CPUs and virtual nodes are properly counted
• Story 2 REPHRASE THIS IN ORDER TO EXPRESS WHAT ADMIN WANT TO DO
  • Card: A group-specific share should be modeled in the same way, regardless the possible local allocation strategies implemented via site config
  • Conversation:
    • a group can be a whole VO or a subset of users part of a VO
    • a group is identified by a set of privilege attributes
    • what kind of relationships are possible between groups and shares?
      • how does this relate to authorization?
  • Acceptance Test:
    • A site admin implements per-group guaranteed share by using a per-group batch queue strategy
    • A site admin implements per-group guaranteed share by using a batch queue with hierarchical share allocation
• Story 3
  • Card: a user wants to know the power available for a certain execution environment
  • Conversation:
  • Acceptance Test: