The members of CASC's Graphs and Irregular Computing (GIC) Group perform research and development across all areas of scalable data science, but particularly on graph problems and others with data-dependent imbalance in their communication features that require nontraditional approaches. Our members contribute to all levels of the mathematics-computing continuum, from pure mathematical analysis and algorithm design to algorithm-hardware co-design to high performance communication, I/O, and accelerator system software libraries to application codes for immediate use by science customers. We provide tooling that enables massive-scale science for scientists and engineers operating across many Lab science and security missions, including but not limited to cyber security, bioinformatics, astronomy and cosmology, materials science, high performance AI, and linear and nonlinear solvers. Many of our members work with the Lab’s Global Security Directorate, and we also receive funding from Laboratory Directed Research and Development, DOE's Office of Science, and other external government agencies.

Group Lead

Min Priest: streaming and sketching algorithms, randomized linear algebra, massive graphs, distributed computing, scientific software

Research Staff

Van Emden Henson: large scale graphs, dynamic graphs, hypergraphs, linear algebra and computational linear algebra applied to data analysis, eigenvalues and eigenvectors, spectral methods in data science, multigrid and algebraic multigrid methods

Marcus Hill: binary analysis, machine learning, vulnerability analysis

Keita Iwabuchi: large-scale graph processing, out-of-core processing, HPC

Tim La Fond: network analysis, dynamic graph algorithms, data mining, anomaly detection

Grace Li: scalable graph algorithms, network analysis, data mining, HPC

Eisha Nathan: network analysis, graph algorithms, dynamic graphs, numerical linear algebra, data mining

Roger Pearce: distributed file systems and tools to profile the I/O performance of data intensive applications

Geoffrey Sanders: algebraic multigrid, eigenspectra, multilinear (tensor) algebra, large-scale graphs

Trevor Steil: parallel graph algorithms, distributed computing, and data science

Andy Yoo: scalable large graph algorithms, large-scale data management, data mining and knowledge discovery, high performance parallel computing, parallel algorithms

Karim Youssef: distributed file systems, memory and storage management, data-intensive application performance tuning, deep learning training I/O optimization