Exploring Emerging Technologies in the HPC Co-Design Space
Staff - Faculty of Informatics
Start date: 5 September 2013
End date: 6 September 2013
The Faculty of Informatics is pleased to announce a seminar given by Jeffrey S. Vetter
DATE: Thursday, September 5th 2013
PLACE: Swiss National Supercomputing Centre CSCS
TIME: 10.00
ABSTRACT:
Recent trends in HPC have forced our community to reexamine the full spectrum of architectures, software, and applications that constitute our HPC ecosystem. Architectural trends, such as heterogeneous processing and nonvolatile memory, are emerging in response to concerns about energy-efficiency and reliability. Meanwhile, applications are being redesigned so that they achieve new scientific objectives, expose prodigious amounts of hierarchical parallelism, and carefully orchestrate data movement. In what we have termed 'co-design,' teams of architects, software designers, and applications scientists, are working collectively to realize an integrated solution. Not surprisingly, this design space can be massive, uncertain, and discontinuous. To assist in this design space exploration, our team has recently developed a number of techniques for modeling, simulating, and measuring these future systems in order to predict performance, power, and reliability. For example, our Aspen (Abstract Scalable Performance Engineering Notation) performance modeling language allows users to compose and solve arbitrary performance modeling questions quickly and rigorously when compared to the traditional approaches.
BIO:
Jeffrey Vetter, Ph.D., holds a joint appointment between Oak Ridge National Laboratory (ORNL) and the Georgia Institute of Technology (GT). At ORNL, Vetter is a Distinguished R&D Staff Member, and the founding group leader of the Future Technologies Group. At GT, Vetter is a Joint Professor in the Computational Science and Engineering School of the College of Computing, and the Principal Investigator of the NSF Keeneland project, a facility for large scale heterogeneous computing using graphics processors. His papers have won awards at the International Parallel and Distributed Processing Symposium and EuroPar; he was awarded the ACM Gordon Bell Prize in 2010. He has recently published an edited book, entitled "Contemporary High Performance Computing: from Petascale toward Exascale," that highlights many, significant HPC systems and sites. For more information, see http://ft.ornl.gov/~vetter/.
HOST: Dr. Sadaf Alam (CSCS), Prof. Olaf Schenk
