A local computing, communications and data-handling company has landed a 4-year $7.6 million grant from the National Science Foundation to develop a system capable of storing and computing enormous amounts of data.
Pittsburgh Supercomputing Center, a University of Pittsburgh and Carnegie Mellon University project, is using the funding to develop a Data Exacell (DXC) prototype.
DXC will update the center’s current “Big Data” handling system called Data Supercell, which is able to hold about 4.2 million gigabytes of information.
The more accessible program will help researchers in fields like biology, astronomy and computer science analyze data at a much faster rate.
Nick Nystrom, director of strategic applications at Pittsburgh Supercomputing Center said DXC will be more than a data storage system.
“The Data Exacell project will include specialized resources to have the words largest shared memory,” he said, “and also special purpose built capabilities for doing graph computations. Those are unique computation resources for doing data analysis that will be part of the Data Exacell pallet.”
The European Bioinformatics Institute now stores about 20 million gigabytes of information on life sciences alone, but access and usability issues plague the industry.
Nystrom said DXC will alleviate those performance roadblocks.
“When people have to do very complicated work flows—going from instruments, say genome sequencers or telescopes or databases, to very large computing resources on which to analyze that data—that’s where the Data Exacell will let them move things more efficiently and make those delays go away.”