Computer Science and Applied Mathematics

The Computer Science and Mathematics department focuses on making sense of large-scale data, including those output by major scientific facilities. To manage such diverse high-volume data, researchers are developing extreme-scale and streaming machine learning capabilities, as well as on-the-wire processing algorithms—all augmented by research on programmingmodels, compilers, performance analysis, modeling/portability, and network protocols.

Projects

• Analysis on the Wire (AoW)
• Automatic Parallelization and Optimization for Lattice QCD Software using a Source-to-Source Compiler
• Exascale Application Development for Lattice Quantum Chromodynamics
• Extreme-Scale Computing for Exascale and Strong-Scaled Problems (XPRESS)
• GPU Research Center
• Integrated End-to-End Performance Prediction and Diagnosis for Extreme Scientific Workflows (IPPD)
• Mesoscale Models for Brittle Fracture of Solids
• Multicore-Aware Data Transfer Middleware: Harnessing Multicore Parallelism to Scale Data Movement
• SOLLVE: Scaling OpenMP with LLVm for Exascale performance and portability
• Software Defined Network Science Flows
• Visualizing and Interacting with Incremental Machine Learning Processes