Leaders of the movement to build universitywide collections of research papers are hoping that cloud computing—a kind of off-site storage for data—will give a new spark to their efforts.
For several years, some librarians have been pushing to build so-called “institutional repositories,” free online collections of papers by their institutions’ researchers. In a new project called DuraCloud, developers are building software to make it easier for librarians to put those repositories in off-site data storage services that are increasingly popular in the commercial sector.
“A key design feature of DuraCloud is to leave the basics of pure storage to those who do it best (storage providers),” said an overview on the project’s Web site.
The project, which got a planning grant from the Andrew W. Mellon Foundation, is now in the pilot phase, but developers hopes to release a version to share with other libraries by the fall of 2010.
A nonprofit group called DuraSpace is spearheading the effort. That group was formed last year by a merger of two groups that produce software to manage digital repositories—the DSpace Foundation, which started at the Massachusetts Institute of Technology, and Fedora Commons, which began at Cornell University.
Carol Minton Morris, a research associate at Cornell who is director of marketing and communications at DuraSpace, said that libraries want more than just storage. The goal of the new service is to add features that support what libraries care about most: access, preservation, re-use and sharing.
“A lot of higher education people talking about the pros and cons of using cloud infrastructure,” she said. The biggest draw of the approach: It can be much cheaper than building new data centers to run on campuses.
One Response to Effort Will Help Libraries Put Academic Papers in Data ‘Cloud’
mbelvadi - April 2, 2010 at 11:02 am
I’d modify that last sentence to “building and maintaining new data centers”. Building it is one thing. Keeping the darn hvac system running reliably and not frying all your data is another, year in and year out, keeping the servers patched against the latest hacker tricks, etc. And new digitization technologies and other research data generating technologies are creating new data that needs reliable stewardship in quantities that are far outstripping available backup methods. Where are the 10TB backup tapes (or other media)? We don’t need all this data to sit on live web servers (cloud or otherwise) with immediate access at all time, but we do need long term storage options that are better than what we have now at any reasonable price.