Information Technology Reference
In-Depth Information
level of service a site is contributing to EGEE; this includes, for instance,
the dei nition of support hours, services run by sites, up-time of services,
mean intervention times, and so on. The EGEE monitoring tools will be
used to enforce these SLAs.
While EGEE's success has been clearly demonstrated, its status as a
series of short-term projects raises the question of the future of the infra-
structure it has built. Already today, many scientii c applications depend
on the EGEE production grid infrastructure, new scientii c collaborations
have been formed thanks to the advancements of grids, and business
and industry are rapidly adopting grid and virtualization technologies.
However, a long-term plan for the availability of the grid service model is
needed to protect the investments of user communities.
As a consequence, EGEE has performed a series of activities together
with national European grid efforts to establish the production grid infra-
structure as a sustained, permanent research tool for science and engi-
neering. The initial outcome of these discussions is an operational model
based on National Grid Infrastructures (NGIs) coordinated by a body ten-
tatively named the European Grid Initiative (EGI). Driven by the needs
and requirements of the research community, it is expected that this setup
will enable the next leap in research infrastructures.
The details of this structure are currently being dei ned by the EGI
Design Study (EGI_DS) project that started in September 2007. Initial con-
siderations have been made in [24] building on the experiences of EGEE
and other national and international grid efforts. Particularly, the NGIs
are expected to provide the building blocks of the infrastructure, much
like the current ROCs in EGEE. However, from an organizational point of
view NGIs are not yet fully operational today. In fact, while most coun-
tries have established NGIs to enable the construction and operation of a
national grid, the diversity between these countries is very high. They
range from single institutes that act as a point of contact for the national
grid community to operational national grid infrastructures; a wide vari-
ety of levels of maturity can be observed.
It is therefore essential that EGI supports the development and progress
of NGIs as much as possible, that EGI dei nes the workload distribution
between the individual NGIs and the EGI together with the appointed
NGI representatives, and that the NGI representatives drive the establish-
ment of the EGI organization. Another contributor to the EGI is the
e-Infrastructure Rel ection Group (eIRG), which has provided the political
roadmap for the establishment of EGI [25,26]. A i rst blueprint of the EGI
structure is expected in summer 2008 and will be available via the EGI_DS
Website (htt p://web.eu-egi.org).
One of the major technical challenges will be the provision of a continu-
ous grid infrastructure during the transition from operations funded by
projects like EGEE and others to the NGI/EGI structure. EGEE is preparing
for that through its decentralized model, which will be further rei ned in
Search WWH ::
Custom Search