High-Performance Computing Department

The High-Performance Computing Department builds students' ability to design and operate massively parallel computing infrastructure, anchored in GPU-accelerated CUDA/OpenCL programming, MPI-based distributed finite element methods, and cluster job scheduling. Students parallelize real large-scale numerical simulation problems - such as seismic modeling - and run and optimize them at petascale. Graduates understand the full hardware-software stack that computation at scale demands and can diagnose and resolve performance bottlenecks in scientific simulations.

⚙️ GeoAcademy System Administration College 🖥️ AI & Computing Sciences College 🔬 Basic Sciences College ⚡ Intelligent Geophysical Exploration College 🛢️ Resource & Energy Engineering College 🌍 Applied Geoscience Solutions College 💼 Economics, Policy & Strategy for the Future College 🎓 Education & Training Development College 🚀 Innovation & International Collaboration College

Machine Learning Department Computer Vision Department Natural Language Processing Department High-Performance Computing Department Quantum Computing Department Generative AI & Foundation Models Center Reinforcement Learning & Agents Center AI Safety & Alignment Center AI for Science Center Robotics & Embodied AI Center

🔑

⭐ Cray

Chair

💡 Supercomputing & Parallel Architecture

Researchers (real GeoVerse Lab members) 14

🧑‍🔬

🔑

1935–2019

Ken Batcher

Researcher

💡 Chief architect of the ILLIAC IV massively parallel computer and inventor of Batcher's sorting networks (bitonic/odd-even merge sort), the foundational parallel-algorithm design techniques for data-parallel HPC systems

🧑‍🔬

🔑

1931–1985

Daniel Slotnick

Researcher

💡 Conceived and led the design of the ILLIAC IV, the first large-scale SIMD (single-instruction-multiple-data) array supercomputer, founding the stream/array-processing architecture that modern GPU computing directly descends from

🔑

1944–2012

Jim Gray

Researcher

💡 Founded the theory of transaction processing and fault-tolerant distributed systems (ACID properties, the transaction concept), directly foundational to reliable distributed-memory HPC job coordination and MPI-based checkpoint/restart systems

🧑‍🔬

🔑

1919–1986

James H. Wilkinson

Researcher

💡 Founded rigorous backward-error analysis for numerical linear algebra algorithms, establishing the stability theory underlying every sparse matrix solver and eigenvalue algorithm used in large-scale scientific computing

🧑‍🔬

🔑

1920–2016

Ray W. Clough

Researcher

💡 Coined the term 'finite element method' and pioneered its application to structural analysis, founding the discretization approach that MPI-based distributed finite element simulations on HPC clusters directly implement at scale

🔑

1926–2019

Fernando J. Corbató

Researcher

💡 Invented time-sharing computer systems (CTSS, Multics), founding the resource-scheduling and multi-user job management concepts that modern HPC cluster schedulers (SLURM, PBS) directly descend from

🔑

1913–2010

Maurice Wilkes

Researcher

💡 Built the EDSAC, one of the first practical stored-program computers, and proposed the concept of cache memory, founding the memory-hierarchy design principle essential to memory-efficient sparse matrix operations on modern HPC systems

🔑

1926–2011

Paul Baran

Researcher

💡 Invented packet switching and designed distributed, redundant network topologies for survivable communication, founding the network-topology design principles underlying HPC cluster interconnects and MPI communication fabrics

🧑‍🔬

🔑

1944–2010

Jean-Claude Laprie

Researcher

💡 Coined the term 'dependability' and founded its systematic taxonomy (reliability, availability, safety, integrity, maintainability), the standard conceptual framework for designing fault-tolerant large-scale HPC systems

🔑

1932–2020

Frances Allen

Researcher

💡 Founded the theory of program optimization and automatic parallelization via control-flow and data-dependence analysis, the compiler theory directly underlying automatic vectorization and parallelization of scientific HPC code

🔑

1937–2018

Larry Roberts

Researcher

💡 Designed and led the ARPANET, the first operational packet-switched wide-area network enabling remote computers to share computing resources, founding the wide-area distributed resource-sharing model that grid and cloud HPC computing directly descend from

🧑‍🔬

🔑

1906–1998

Reynold B. Johnson

Researcher

💡 Invented the hard disk drive (IBM RAMAC), founding random-access magnetic storage technology that all modern parallel file systems and distributed storage architectures for HPC I/O are built upon

🔑

1932–2007

Gene H. Golub

Researcher

💡 Founded modern matrix computation theory including the golden-standard SVD algorithm and iterative solver methods, foundational to the large-scale sparse linear system solvers used throughout HPC scientific simulation

🧑‍🔬

🔑

1925–2002

John Cocke

Researcher

💡 Pioneered RISC (Reduced Instruction Set Computer) architecture and instruction pipelining, founding the processor design philosophy underlying the high clock-rate, high-throughput CPU/GPU cores used throughout modern HPC hardware