Oak Ridge Leadership Computing Facility

Scientific Computing

The Scientific Computing Group (SCG) works concurrently with the users of HPC systems to help them obtain optimal results from the OLCF’s computational resources. The SCG is comprised of research scientists, visualization specialists, and work-flow experts who are trained in chemistry, physics, astrophysics, mathematics, numerical analysis, or computer science. Each research team using an OLCF system is assigned an SCG liaison who is familiar with the field of research. SCG liaisons actively participate in the research, help design and optimize code for the users’ applications, streamline the work flow, and solve any computer issues that arise. Visualization specialists capture the resulting data in images and help the users analyze it. Contact Ricky A. Kendall, group leader, at kendallra@ornl.gov.

The User Assistance and Outreach (UAO) Group provides support to the OLCF users, mediation between the outside world and the OLCF, and acquaints the public with the work conducted at the OLCF. The group creates accounts for news users, provides Level I and Level II technical support to research teams, and generates documentation on OLCF systems access, policies, and procedures. UAO staff members attend the meetings of the other OLCF groups, communicating suggestions from users and participating in decisions on the users’ behalf. The group also creates science research highlights, writes SciDAC Review and HPCwire articles, produces podcasts for scientists and the public, and connects the OLCF with universities across the nation. Contact Ashley D. Barker, group leader, at ashley@ornl.gov.

The Technology Integration (TechInt) group is charged with delivering new technologies into the Leadership Computing Facility (OLCF) and other projects at ORNL by identifying gaps in the system software stack and working with the R&D community to develop, harden, and deploy solutions. Their technology scope includes archival storage, parallel file systems, data management, high-performance networking, and architecture. TechInt staff work behind the scenes to develop the infrastructure that supports the NCCS systems and keep it ahead of the technology curve. They research and evaluate emerging technologies in the aforementioned areas and provide the systems programming to seamlessly integrate technologies and tools into the infrastructure as they are adopted. Contact Sudharshan Vazhkudai, group leader, at vazhkudaiss@ornl.gov.

The High Performance Computing Operations (HPCO) Group keeps the OLCF leadership supercomputing systems running. Members of the group monitor all systems 24 hours a day, seven days a week, 365 days a year, and are responsible for administration, configuration management, and cybersecurity. The staff works with infrastructure systems, with Titan, and with other OLCF supercomputers. The HPCO tests the systems when they are installed and upgraded, and use diagnostic tools to continually monitor them. They anticipate problems before they arise and identify components that are near failure. The group also ensures that all systems conform to ORNL cybersecurity policy. Contact Kevin Thach, group leader, at kthach@ornl.gov.

The Application Performance Tools Group researches, tracks, and purchases a wide range of software tools that help science researchers access and improve the performance of their applications on current and emerging OLCF computing systems. The group also manages the contacts with the vendors for the purchase of new modeling tools, languages, middleware, and performance-characterization tools. The group primarily focuses on issues that arise for research applications when they are run on very large-scale systems, such as the OLCF’s Titan supercomputer. Contact Rich Graham, group leader, at rlgraham@ornl.gov.