Main Page
REDDnet: Enabling Data Intensive Science in the Wide Area
REDDnet (Research and Education Data Depot network) is an NSF-funded project to provide a large distributed facility with a core focus of data intensive collaboration in the wide area Our mission is to provide working storage, not archival or long-term personal storage, for research collaborations who are either trying to move data from one collaborating institution to another or to allow these collaborations to share large data sets for brief periods of time (a few months) while they work on it.
One example comes from the CMS collaboration, a high energy physics experiment that will be taking data soon at the Large Hadron Collider (LHC) at CERN. Groups of researchers, distributed across the country and the world, will want to use data products derived from the raw data produced by collisions in the LHC to do a variety of tasks from calibrating the detector to searching for new physics. They will want these data products available for anywhere from a month to a few months, and they would benefit greatly if this data could be available for processing on their distributed computing infrastructure and in particular on the Open Science Grid. Another example, from the AmericaView project, might occur in the aftermath of an earthquake in California or a Hurricane on the Gulf Coast, where researchers across the country will want access to the geospatial image data from satellites covering the affected region. For a few months after the event, this data could be uploaded to REDDnet and made available to this community. Initially, REDDnet will deploy >700 Terabytes of distributed storage with an emphasis on scalability, speed and fault tolerance. Currently (Spring 08), there are roughly 160 TB deployed. For example, at the Supercomputing 2006 Conference in Tampa, Florida, REDDnet demonstrated sustained transfers at a rate of 10 Gigabits per second between Caltech and the convention floor. These transfers were limited by the bandwidth of the network connection. At the same conference, REDDnet demonstrated fault tolerance by striping data across thirty depots and then successfully reading the data even after turning off nine of these depots. |
Research Projects Using REDDnet
- AmericaView - Satellite remote sensing data and technologies in support of applied research, K-16 education, workforce development, and technology transfer.
- Structural Biology - Image reconstruction of large macromolecular assemblies through a collaborative effort of Vanderbilt and Lawrence Berkeley National Laboratory researchers.
- Terascale Supernova Initiative - a multidisciplinary collaboration to develop models for core collapse supernovae and related enabling technologies.
- National Geospatial Digital Archive (NGDA) - a collecting network for the archiving of geospatial images and data.
- Retinopathy - Diabetic Eye Disease Screening in Peru and Bolivia
Collaborators
Core Institutions
Error creating thumbnail: Unable to save thumbnail to destination |
Error creating thumbnail: Unable to save thumbnail to destination |
Error creating thumbnail: Unable to save thumbnail to destination |
||||
Vanderbilt | Tennessee | S. F. Austin | ORNL | Nevoa Networks | N. C. State | Delaware |
Collaborating Host Institutions
São Paulo | Rio de Janeiro | Michigan | Florida | Fermilab | Caltech |
Error creating thumbnail: Unable to save thumbnail to destination |
|||||
AMPATH | FIU | Library of Congress | SDSC | Stanford | UCSB |
Support
This work is supported by NSF Grant PHY-0619847 and by the Vanderbilt Center for the Americas