Today scientific collaborations deal with a variety of large-scale trans-continental projects in the domain of computing. This presents a number of interesting challenges, covering not only the scale and distributed nature of the data sets or the complexity of the processing applications in use; but also the different underlying compute management technologies used by participating infrastructures.
As a software platform for high performance distributed computing EMI, the European Middleware Initiative, represents the response to all these interesting challenges, and, thanks to its releases, has become something that most of the European experts can depend on for day-to-day work within their respective collaborations. In particular, EMI is a crucial element of Grid computing infrastructure - the Worldwide LHC Computing Grid (WLCG) - designed with the main focus of solving the large challenges of the particle physics Large Hadron Collider (LHC) experiments at CERN. This infrastructure allows access to an enormous amount of distributed computing resources and archives, and is able to support powerful, complicated and time-consuming data analysis.
“The EMI collaboration is proud to seethat all the Grid software components they had developed in the previous years, with the support of the European Commission and many European research institutions, are successfully sustaining the load of thousands of researchers working every day”, says Alberto Di Meglio, EMI project director.
EMI is a joint effort of the major European distributed computing middleware providers (ARC, dCache, gLite, and UNICORE). This initiative introduces in Europe and beyond new functionality to satisfy user requirements specifically in four sectors: security, compute, data and information systems. These sectors are working together to supply products that can be installed from a distribution source by site administrators or researchers. This supports and evolves the research infrastructures and allows academic and industrial researchers to access resources, data and applications across the world, targeting the needs of users with a huge computational and storage demand.
EMI improves the existing middleware services and harmonizes them, realizing a common framework with the result of rendering the middleware to be simpler and easier to use, reducing at the same time the interoperability problems faced by the distributed computing infrastructure providers. Thanks to EMI, it was possible to get the results in a previously unthinkable time and to contribute to the discovery of one of the most ‘wanted’ particles: a Higgs boson.
“Without the EMI middleware, such an important result could not have been achieved in such a short time – concludes Alberto - EMI was instrumental in allowing scientists to work at the same time in a transparent way on hundreds of computing centres located in many different countries, include in their results the last collected data and obtain sufficient statistics (thousands of trillions of high energy proton-proton collisions in 2011 and 2012) to be able to discover what they are confident to be a Higgs boson.”
In parallel to the LHC experiments, the EMI project supports a number of other initiatives.
To sustain the creation of an open data e-infrastructure for the conservation of marine living resources, EMI formalized a partnership with iMarine for the adoption of EMI services through their data exploitation within gCube software. Furthermore, together with EGI’s Life Science Computing Grid (LSCG), EMI is also involved in LINGA(LInked Neuroscientific Grand chAllenge), the first data and compute challenge involving neuro-scientific infrastructures from two continents (neuGRID in Europe with CBRAINand LONI, Laboratory of Neuro Imaging in North America). Some 11,000 brain scans from distributed neuro-centers have been already analyzed using computationally demanding image processing applications for marker discovery in neuro-degenerative diseases like Alzheimer’s.
These two examples are a good demonstration of the stability and usability of EMI services for large-scale scientific research.
To know more:
· ARC offers compliant, portable, platform independent, open source middleware. The community using it is the unique distributed Nordic Tier-1 for CERN LHC experiments ATLAS and ALICE.
· dCache is a storage solution for huge amounts of data. Over 80 deployments use it within the WLCG. Some provides many petabytes (PB) storage capacity for local users. Its deployments lead some 60PB storage capacity to CERN's LHC experiments.
· gLite,the next generation middleware for distributed Grid computing, was born as part of the Enabling Grids for E-Science in Europe (EGEE) projects, which provides a robust framework of services whose components satisfy the requirements of the CERN’sLHCexperiments need.
· UNICORE, initiated in the supercomputing domain, today is a general-purpose Grid technology. Its recent version, UNICORE 6 follows the latest standards from the Grid and Web Services world and offers a rich set of features to its users.