Enabling Grids for E-science
EGEE (Enabling Grids for E-Science ) is a project financed in the 7th Framework Program of the European Union by the Directorate-F: Emerging Technologies and Infrastructures. In its third phase from 2008 to 2010, it connects more than 260 institutions in 55 countries to create a research grid infrastructure for the European Research Area .
Scientists from research and teaching as well as users from industry are offered access to large computing and storage capacities - regardless of the geographical location. Current research results in the field of grid technologies are used to develop a service grid infrastructure, which is made available 24 hours a day. With funding of over 30 million euros per phase, the EGEE project is one of the most expensive of its kind. The EGEE project also focuses on attracting a large number of new users.
The main focus of the project is on the following areas:
- Construction of a consistent, robust and secure grid network, which attracts additional computing power
- Development of a middleware gLite, especially for use in various scientific disciplines
- Maintain and continuously improve the middleware to provide reliable services to users
- Acquiring new users from different scientific domains
- Provision of required training and support at a high level.
background
EGEE began in April 2004 under the name Enabling Grids for E-science in Europe and shortly thereafter changed its name to Enabling Grids for E-science, after partners from the USA and Southeast Asia had also joined the project.
The starting point for EGEE was the already existing LHC Computing Grid (LCG). The aim of LCG is to create resources for the expected amounts of data from the Large Hadron Collider (LHC) at Geneva’s CERN . LCG connects high-energy physics computing nodes around the world in order to be able to cope with the data volume of the projected 15 petabytes per year produced by the LHC. Starting from LCG, EGEE started to combine further resources from all over the world and above all other user groups and applications to form the largest multidisciplinary production grid.
Middleware
EGEE started its work on the basis of the LCG-2 middleware of the LCG project. This in turn was based on the developments of the previous EGEE project DataGrid. At the same time, the development of gLite was started and different sources were used. gLite is a lightweight middleware that offers all basic grid services.
In March 2009 gLite arrived at version 3.2 and now consists of several hundred software packages that are grouped into logical installation modules.
The gLite middleware is also used by a number of groups outside of EGEE, such as the EU-funded DILLIGENT project. The French space organization CNES is also planning to use gLite in the future.
The gLite middleware was designed for GNU / Linux on the i386 architecture. The recommended distribution is Scientific Linux , but it is widely operated under other RHEL -compatible distributions ( e.g. CentOS ). The main development branch of the middleware is now intended for operation under Scientific Linux 5 x86_64 and Debian GNU / Linux 4. With a few exceptions, the most important components of this version are already certified for operation (as of March 2010). The remaining components are being ported.
Infrastructure
EGEE now operates almost 25 data centers from research institutions, universities, companies or other interested participants. These were originally only found in Europe, but now there are also nodes in the USA and Southeast Asia. The grid nodes in EGEE meanwhile provide around 110,000 CPU cores.
What does the future hold?
EGEE ended as a project in March 2006. The EU follow-up project EGEE II started on April 1, 2006 and ran until the end of April 2008. The two-year follow-up project EGEE-III has been running since May 1, 2008.
EGEE II continued the results of EGEE. EGEE was planned as the first two-year phase of a four-year program to offer a production-quality grid infrastructure for the European research landscape and beyond. Scientists in universities and industry are already benefiting from the EGEE e-infrastructure, which supports many applications from different scientific disciplines at the same time around the clock. It enables the use of a common resource pool regardless of the geographic location with access to important storage, computer and network facilities. The EGEE II project significantly expanded and consolidated this infrastructure. It incorporates national, regional and thematic grid initiatives and works with other grids around the world. The high capacity resulting from the global infrastructure by far exceeds the capacity of local clusters and individual centers. The result is a single tool for collaborative, computationally intensive science (“e-science”). EGEE II also expanded the portfolio of supported applications to more than 15 disciplines and this number is constantly increasing.
EGEE III is the logical successor to EGEE II and aims to expand the infrastructure and areas of application, as well as to create a sustainable European grid infrastructure.