Feature - Open Science Grid: Living up to its Name
The Open Science Grid, supported by the U.S. Department of Energy’s Office of Science and the National Science Foundation, works hard to live up to its name: OSG is a grass roots consortium open to all scientific communities, accommodating to all resource providers, open to new technologies and to integration and interoperability with other grids. OSG supports science and research—especially collaborative science requiring high throughput computing. OSG enables the sharing of distributed computing and storage resources by connecting them into a common grid infrastructure over production and research networks via a common set of middleware.
OSG’s mission is to help satisfy the ever-growing computing and data management requirements of scientific researchers.
The OSG is a unique alliance of universities, national laboratories, scientific collaborations and software developers. Members’ independently owned and managed resources make up the distributed facility, agreements between them provide the glue for it, their requirements drive its evolution, and their effort makes it happen. OSG provides a ‘virtual facility’ onto which individual research communities, the ‘virtual organizations,’ add services according to their scientists’ needs.
A number of physics collaborations provided the initial driving force for OSG and now rely on it for their data processing. These include the high energy physics experiments at CERN’s Large Hadron Collider, the STAR nuclear physics experiment, the Laser Interferometer Gravitational Wave Observatory and the current experiments at Fermilab’s Tevatron. OSG has grown to also support other branches of physics, biology and mathematics.
“Many of these physics experiments must sift out discoveries from prodigious quantities of data produced by detectors with up to hundreds of millions of electronics channels,” says Torre Wenaus, applications co-coordinator for OSG. “At the LHC in particular, the unprecedented particle collision energies will produce very complex data events. High complexity plus high rate requires vast amounts of processing power.”
OSG works with an expanding set of research communities to help them evaluate their cyberinfrastructure needs and plan their solutions both locally across the campus and as part of national or international efforts. OSG also organizes hands-on training schools and education activities to help students and teachers learn how to use and provide distributed computing.
The OSG software is based on the Virtual Data Toolkit—a packaging and distribution of Condor, Globus and other software needed to support a distributed facility. The OSG software stack is installed on participating grid computing nodes and includes a client package for end-user scientists. The OSG provides a test environment for integrating new services and functionality and a fully supported distributed production environment for running applications.
OSG members work actively with partners and peers, in particular Enabling Grids for E-sciencE in Europe and TeraGrid in the United States, network organizations such as ESnet and Internet2, as well as campus, regional, national and international grids to achieve interoperability.
“We want to open up gateways to enable communication between grids so that scientists can run the same application on multiple grids,” says Alan Blatecky, OSG’s engagement coordinator. “Interoperability increases the resources available to individual scientists, allowing them to become more productive.”
To learn more visit the Open Science Grid Web site.
- Anne Heavey, OSG