Towards enabling high user productivity and high-performance with parallel computing
Recent trends in computer architecture are making high degrees of parallelism as well as heterogeneity ubiquitous. This creates significant challenges to application developers as well as compiler implementations. The effort to develop parallel applications for advanced computational models is extremely high. Further, currently it is impossible to achieve performance portability of high-performance applications from a single version of a program – different code versions are necessary for different target platforms, e.g., for multicore CPUs versus GPUs. A promising approach to easing the burden on applications programmers and achieving high performance on parallel machines is via identifying suitable high-level and domain-specific abstractions. This talk will discuss efforts to develop compiler techniques to automatically transform programs specified using high-level abstractions.
About the Speaker:
Dr. P. Sadayappan is a Professor in the Department of Computer Science and Engineering at The Ohio State University. His primary research interests center around performance optimization and compiler/ runtime systems for high-performance computing, with special emphasis on high-performance frameworks that enable high productivity for application developers in scientific computing. Two recent projects include a polyhedral framework for automatic parallelization and data locality optimization, and the Tensor Contraction Engine – a domain-specific compiler/runtime system to automatically transform high-level specifications into efficient parallel programs, for a class of high accuracy ab initio models in quantum chemistry. Dr. P. Sadayappan obtained a B.Tech from the Indian Institute of Technology, Madras, and M.S. and Ph.D. from Stony Brook University, all in Electrical Engineering.
Faculty Host:Prof. Somnath Ghosh, 203 Latrobe, 410-516-7833, [email protected]
For more information, please contact Jae Hong, 410-516-5033, [email protected]
Dental Enamel- a Multi-Scale Modelling Challenge
The tooth is a unique functionally graded composite structure at several levels providing a hard and apparently self-healing enamel external shell bonded to a dynamic and resilient dentin core both supported by a vascular and neural network in the tooth pulp. Tooth enamel is nature’s cell derived method for production of a high elastic modulus (~ 90 GPa), hard, wear, and fatigue resistant structure. This presentation will review the micro and meso structure of human teeth as well as studies on their Hertzian contact and Vickers indentation response. The fracture toughness measurements of enamel and dentin by several groups and the need to further explore mechanical response with enamel location and orientation are discussed. Emphasis well be on the role of decussation on enamel properties and likely mechanisms for enamel self-repair of microcracks when teeth are fatigued
About the Speaker
Van P. Thompson, DDS, PhD, is currently, Professor of Biomaterials, Biomimetics and Biophotonics at King’s College London Dental Institute and was previously Chair, Biomaterials and Biomimetics, NYU College of Dentistry. Known for his work on adhesion and bonded bridges at the University of Maryland he has published many articles and made numerous presentations on dental biomaterials in the U.S. and internationally. His current research areas include dentin caries activity, all-ceramic crown fatigue and fracture, modifications of dentin for bonding, engineering tissue response via scaffold architecture and practice based research (PEARL Network).
Faculty Host: Prof. Somnath Ghosh, 203 Latrobe, 410-516-7833, [email protected]
Khairul Bariah Abd Majid: 410-516-5033 or [email protected]