Living cyberinfrastructures and their impact on scientific research
Project Leads: Rich Knepper, Campus Bridging and Research Infrastructure, UITS Research Technologies
Research made possible by High Performance Systems & Scientific Applications and Performance Tuning, UITS Research Technologies
RT Manager for Campus Bridging and Research Infrastructure Rich Knepper presented a talk at the International Conference on Infrastructures and Cooperation in E-science and E-humanities, Leipzig, Germany, in June 2014 titled “Living cyberinfrastructures and their impact on scientific research.” He explored the evolving model of center-based cyberinfrastructure and some of the challenges that attend that model. The presentation is available on YouTube at https://www.youtube.com/watch?v=LQygr2p-EPM
There are multiple impacts of living cyberinfrastructure (CI) on researchers and the science/research process itself. Free cycles mean researchers can accomplish a lot of work. In exchange, to get time on the resource, they may be required to fit into the resource’s computational model, take part in an allocations process, and justify their science. This can cut down on early experimentation. They may have to adapt their software to work on the system, share the system (and possibly their data) with other researchers, agree to provide attribution for the center, or participate in surveys or other feedback mechanisms to help centers understand how things are going. Centers themselves face challenges. They exist in a cooperative-competitive tension, lack adequate funding, face scaling issues that come with exascale computing, and need to attract larger communities. Thus, living CI can claim part of the scientific process.
Living CI implies a close link between technology and scientific outcomes: Science becomes part of the technology that supports it. As dedicated single-scientist resources grow less feasible, cyberinfrastructure based on computation centers like XSEDE present options for conducting science. Scientists access free cycles, but face the tradeoffs noted above. Centers face their own challenges, including what Knepper calls “inertia of purpose,” attracting larger communities of users, funding, and the move to exascale. Centers need a new sustainability model beyond collaboration and competition. Knepper discusses questions about the relationship between science and living CI, and suggests examples of CI designed for new disciplines–a networked opera, a computer-controlled light tower, or the Center for Computational Social Sciences at Stanford.
The Campus Bridging and Research Infrastructure (CBRI) group enables IU researchers, scholars, and artists to bridge easily and effectively from their personal and lab computing systems to IU’s advanced research cyberinfrastructure to national cyberinfrastructure resources such as XSEDE (the eXtreme Science and Engineering Discovery Environment).
The High Performance Systems (HPS) group implements, operates, and supports some of the fastest supercomputers in the world – IU’s Big Red II, the Quarry cluster, and the large memory Mason system – in order to advance Indiana University's mission in research, training, and engagement in the state. HPS also supports databases and database engines used by the IU community.
The Scientific Applications and Performance Tuning (SciAPT) group delivers and supports software tools that promote effective and efficient use of IU’s advanced cyberinfrastructure – which, in turn, improves research and enables discoveries.
NSF GSS Codes:
Primary Field: Multidisciplinary Studies (980) - Science, Technology, and Society
Secondary Field: Computer Science (401) - Information Science/Studies