Intro

Title: International cooperation with the USA on common e-Infrastructure for scientific data

• e-Infrastructure: Grid/Cloud/Super Computing
• Need USA partner -- need support from NSF

The objective is

to establish an EU/USA coordination platform aiming at full interoperability of scientific data infrastructures, and

• EU/USA coordination platform
  • main goal is the organisation (the forum) to discuss interoperability of SDI's
  • the forum should result in a proposed standard(s), reference implementation, and 3-4 applications which use the reference implementation
• scientific data infrastructures:
  • data access & movement, metadata catalogues, data replica catalogs

to demonstrate this coordination through several joint EU-USA prototypes that would ensure persistent availability and effective sharing of data across scientific domains, organisations and national boundaries.

• several (== 3 or 4) joint EU-USA prototypes. Ideas of potential prototypes from HEP:
  • Remote data access from job (Xrootd, http across the WAN)
  • Common Cloud storage API: Amazon S3, OpenStack...
  • ...

The platform should provide for:

the collection of requirements and approaches for standardisation (development, promotion, adoption and maintenance);

• Workshop to collect requirements, see various scientific data infrastructures
• Form Technical Working Group:
  • Requirements deliverable document
  • prepare solution approaches before next workshop

common ICT infrastructure approaches (technical, semantic, reference architecture, financing models, etc) in order to lower access barriers;

• Workshop to discuss proposed solutions -- technical proposals for interoperability between SDI's

harmonisation of intellectual property frameworks for scientific information;

• Open access licences for all technical deliverables
• Papers published in peer reviewed conferences/journals

and mechanisms for international networking of experts and multidisciplinary communities.

• Need 2 or 4 workshops (1/2 EU, 1/2 USA)
  • 1st to discuss requirements, catalog existing scientific data infrastructures, networking
    • Form Technical Working Group
  • 2nd where parties present proposals for prototype common data infrastructure
  • 3rd reference implementation
  • 4th applications using the reference implementation.

• Workplans:
  • NA3.2.1: Project Management
  • NA3.2.2: Coordination
    • Organize workshops (2 of them. USA should organise the other half).
    • Disseminations, Training.

The joint prototypes should leverage and build upon similar initiatives in Europe and USA.

• Similar initiatives:
  • WLCG storage
  • SRM
  • WLCG Storage Jamboree: Dynamic data dist., Xrootd, NFS 4.1, HTTP, Data caching,
  • Cloud Computing R&D: search for common data API: Amazon S3, OCCI

The proposal should clearly describe synergies and collaboration with corresponding existing or potential NSF-funded initiatives.

• Main collab should be with OSG
• Also specific HEP contacts -- ATLAS, CMS NSF-funded projects?

-- DanielVanDerSter - 18-Oct-2011