iSGTW - International Science Grid This Week
iSGTW - International Science Grid This Week

Home > iSGTW 03 October 2007 > iSGTW Resources - Choose and start to use your grid

Learn: Choose and start to use your grid

A few of the many grid projects available.
Image courtesy of NorduGrid and Vicky White

Grid technology continues to improve and new grids and grid projects are appearing across the planet. From campus grids to massive international projects, opportunities to get involved with grids and e-science are growing.

But how do you get started? What jobs do the different grids support? Who can join? Will you need a computing degree to work it all out?

This article provides a quick snapshot of four very different grid infrastructures—Enabling Grids for E-sciencE, Open Science Grid, GridPP and TeraGrid—and provides an overview of what they’re doing and how you can get involved.

Enabling Grids for E-sciencE

Open Science Grid




Enabling Grids for E-SciencE

What do you do?

The Enabling Grids for E-sciencE project provides scientists and engineers from 48 countries with a seamless grid infrastructure for e-science, available 24 hours a day.

What is EGEE’s capacity?

EGEE has over 41,000 CPUs at 240 international sites constantly available to more than 5000 users. The EGEE infrastructure runs in excess of 100,000 jobs a day and can store more than five million Gigabytes of data.

What jobs do you support?

EGEE is ideal for a wide variety of scientific research, from particle physics to the life sciences, especially where the time and resources required render applications impractical. EGEE supports MPI-based jobs on a limited number of sites and plans to expand this in the future.

What middleware and software do you use?

EGEE runs the gLite middleware, reengineered by the project from a range of open source middleware solutions. gLite runs on Scientific Linux and is being ported to other platforms.

Can I access EGEE resources?

Anyone has the potential to access or contribute to EGEE resources: you need a digital certificate and a Virtual Organization to join. EGEE is fully committed to supporting the maximum range of research domains and applications, and as a result is always keen for others to participate in its work. You can get involved as an end user, the manager of a VO or as a resource provider.

How do I get started?

You can obtain your digital certificate from your local certification authority: search for your region from the Asia Pacific, Europe or the Americas. You can then browse through EGEE’s existing applications to find a community you’d like to join. For more information visit the User and Application Portal or register as a member of the EGEE community.

What does the future hold for EGEE?

The EGEE grid is expected to grow in capacity considerably in the run up to the Large Hadron Collider start-up in 2008, and will also continue to add new communities as it moves into a third proposed phase next year.

Who’s backing this project?

EGEE is supported by the European Union and a collaboration of more than 90 organisations from over 30 countries.

- Owen Appleton, EGEE



Open Science Grid

What do you do?

Open Science Grid offers a shared infrastructure of distributed computing resources, independently owned and managed by its members. Together, OSG members provide a virtual facility available to individual research communities, who can add services according to their scientists’ needs. OSG delivers the U.S.-based infrastructure to support the Large Hadron Collider experiments.

What is OSG’s capacity?

OSG currently has processing and storage resources accessible from more than 50 institutions, including self-operated research Virtual Organizations, campus grids, regional grids and OSG-operated VOs. The OSG is providing about 10,000 CPU days per day in processing and 10 Terabytes per day in data transport. OSG CPU usage is at about 75%.

What jobs do you support?

OSG suits jobs requiring high-throughput processing and data movement. Some OSG sites are now supporting MPI jobs and support for this will be increased in the future. OSG jobs currently come from fields including high energy, nuclear, astro and gravitational wave physics, as well as nanotechnology and bioinformatics.

What middleware and software do you use?

OSG’s Virtual Data Toolkit provides packaged, tested and supported collections of middleware for installation on participating compute and storage nodes and a client package for end-user researchers. The package includes Condor, Globus, dCache, Authz, VOMS, GUMS and Gratia. Individual Virtual Organizations can also build their own applications to run on OSG.

Can I access OSG resources?

The OSG Consortium is open to any research or scientific community who wants to participate and contribute. Individual OSG sites set their own policies regarding which OSG members can access their resources. Some sites restrict their resources to a limited set of VOs; others make them available for opportunistic use by all OSG-registered VOs.

How do I get started?

You can start using OSG by joining or forming a VO, registering with OSG, then downloading and installing the OSG software package. At this point you can begin to run jobs and/or register resources. Please contact OSG for more information.

What does the future hold for OSG?

OSG members direct OSG’s evolution according to their requirements. To this end, OSG plans to continue to expand its reach, capacity and services to meet the evolving needs of its stakeholders.

Who’s backing this project?

The Open Science Grid is supported by the U.S. Department of Energy’s Office of Science and the National Science Foundation.

- Anne Heavey, OSG




What do you do?

GridPP is the UK grid for particle physics. The main focus for GridPP is the provision of computing for the Large Hadron Collider experiments, but we are also fully integrated into EGEE and through this support many different sciences.

What is GridPP’s capacity?

GridPP integrates CPU and storage resources from across 17 UK institutes. GridPP currently have about 9000 processors and over 1000 Terabytes of disk. We have over 3000 users within GridPP-supported Virtual Organizations, and somewhere between 500 and 1500 users regularly submit jobs.

What jobs do you support?

The GridPP infrastructure is most suited to highly intensive non-parallel jobs with large input/output demands. GridPP is also used for data storage, movement and processing. MPI is not widely supported, but is now possible and implemented at some GridPP sites, while jobs generating lots of small files can be challenging. The infrastructure is not currently well adapted for secure processing and data storage, but work is ongoing within EGEE to provide finer support to Virtual Organizations.

What middleware and software do you use?

GridPP is part of EGEE and as such uses the gLite middleware. The application software depends on the Virtual Organization.

Can I access GridPP resources?

If you are a scientist working at a UK-based institution, you are welcome to start using GridPP. As a grid for particle physics, GridPP provides preferential access to UK high energy physics collaborations, but a minimum of one percent of resources is allocated to international EGEE Virtual Organization usage, with individual GridPP sites free to decide whether they will support a given VO. In fact, most large active EGEE VOs are now enabled on GridPP sites and use much greater than one percent of GridPP resources. However, once the LHC starts this usage will be more restricted.

How do I get started?

You can find out more about GridPP at our Web site. The best way to get started is to contact someone at or near your institute who already uses GridPP or EGEE and can guide you through. As a first step, please contact GridPP’s Dissemination Officer, Sarah Pearce, who will put you in touch with experts near you.

What does the future hold for GridPP?

GridPP is growing rapidly, in preparation for the LHC. The largest GridPP site will scale by a factor of four over the coming years. Growth in the other centers will depend on the funding they receive: there are several universities where access to campus grids and HPC facilities may greatly increase individual site contributions.

Who’s backing this project?

GridPP is funded by the UK government through the Science and Technology Facilities Council as part of its e-Science Programme.

- Sarah Pearce, GridPP




What do you do?

TeraGrid provides a U.S.-wide high performance computing and communications grid infrastructure, coordinated by the Grid Infrastructure Group at the University of Chicago in partnership with nine resource-provider sites: Indiana University, Oak Ridge National Laboratory, the National Center for Supercomputing Applications, Pittsburgh Supercomputing Center, Purdue University, San Diego Supercomputer Center, Texas Advanced Computing Center, University of Chicago/Argonne National Laboratory and the National Center for Atmospheric Research.

What is TeraGrid’s capacity?

Our current resources include 250 teraflops of computing capacity and 30 petabytes of data storage and systems. About 4,000 scientists and engineers from a wide range of disciplines run applications using Teragrid.

What jobs do you support?

TeraGrid resources support complex workflows for very large-scale science and engineering research, including MPI jobs that can run across thousands of cores. Our large shared-memory machines also enable popular commercial and community codes for simulation, data mining and visualization.

What middleware and software do you use?

TeraGrid’s Coordinated TeraGrid Software and Services is built on Globus Toolkit 4, Condor and other middleware components. A range of scientific application packages spanning the disciplines are also available for use by the research community.

Can I access TeraGrid resources?

TeraGrid resources are available to researchers and educators at U.S. academic or non-profit research institutions, and to their collaborators at national and international institutions. TeraGrid Science Gateways can then provide your research community with access to resources tailored to your needs.

How do I get started?

You can apply on-line for an allocation of TeraGrid resources at no cost. Applications from all fields of research are considered. Processing time varies with size: start-up accounts can be provided within two weeks; medium-sized accounts are provided each quarter; and allocations running to millions of computing hours are provided twice a year.

What does the future hold for TeraGrid?

The National Science Foundation is making substantial investments in high performance computing resources. Planned additions to TeraGrid include a 500-teraflop system in 2008 and a petaflop system in 2009. Our total capacity will be well into the peta-scale range by 2011.

Who’s backing this project?

TeraGrid is sponsored by the U.S. National Science Foundation Office of Cyberinfrastructure.

- Michael Schneider, TeraGrid




 iSGTW 22 December 2010

Feature – Army of Women allies with CaBIG for online longitudinal studies

Special Announcement - iSGTW on Holiday

Video of the Week - Learn about LiDAR


NeHC launches social media

PRACE announces third Tier-0 machine

iRODS 2011 User Group Meeting

Jobs in distributed computing


Enter your email address to subscribe to iSGTW.


 iSGTW Blog Watch

Keep up with the grid’s blogosphere

 Mark your calendar

December 2010

13-18, AGU Fall Meeting

14-16, UCC 2010

17, ICETI 2011 and ICSIT 2011

24, Abstract Submission deadline, EGI User Forum


January 2011

11, HPCS 2011 Submission Deadline

11, SPCloud 2011

22, ALENEX11

30 Jan – 3 Feb, ESCC/Internet2


February 2011

1 - 4, GlobusWorld '11

2, Lift 11

15 - 16, Cloudscape III

More calendar items . . .


FooterINFSOMEuropean CommissionDepartment of EnergyNational¬†Science¬†Foundation RSSHeadlines | Site Map