GeoVerse Lab
โ† AI & Computing Sciences Division

High-Performance Computing Center

Provides massively parallel computing infrastructure - GPU-accelerated CUDA/OpenCL seismic modeling, MPI-based distributed finite element methods, and cluster scheduling - enabling petascale computations for large-scale numerical simulation.

Machine Learning CenterNatural Language Processing CenterComputer Vision CenterHigh-Performance Computing CenterQuantum Computing CenterGenerative AI & Foundation Models CenterReinforcement Learning & Agents CenterAI Safety & Alignment CenterAI for Science CenterRobotics & Embodied AI Center
Research Fields 14
Parallel Algorithm Design & Sorting Networks
๐Ÿ”€ Parallel Algorithm Design & Sorting Networks
A sorting network is a fixed sequence of compare-and-swap operations between pairs of posiโ€ฆ
SIMD Array Processing & Stream Computing Architecture
๐Ÿงฎ SIMD Array Processing & Stream Computing Architecture
A single-instruction, multiple-data architecture broadcasts one shared instruction stream โ€ฆ
Distributed Computing & Fault-Tolerant Transaction Processing
๐Ÿ’พ Distributed Computing & Fault-Tolerant Transaction Processing
A distributed transaction is a unit of work spanning multiple separate machines that must โ€ฆ
Numerical Linear Algebra & Sparse Matrix Methods
๐Ÿ“ Numerical Linear Algebra & Sparse Matrix Methods
Numerical linear algebra studies how to solve linear-algebra problems, such as solving a sโ€ฆ
Finite Element Methods for Large-Scale Simulation
๐Ÿ—๏ธ Finite Element Methods for Large-Scale Simulation
The finite element method solves a partial differential equation governing a continuous phโ€ฆ
Job Scheduling & Time-Sharing Resource Management
โฑ๏ธ Job Scheduling & Time-Sharing Resource Management
A time-sharing scheduler divides a single computer's processing time into short slices andโ€ฆ
Memory Hierarchy & Cache Optimization
๐Ÿ—„๏ธ Memory Hierarchy & Cache Optimization
A memory hierarchy arranges several tiers of storage with different speeds, sizes and costโ€ฆ
Interconnection Networks & Distributed Network Topology
๐Ÿ•ธ๏ธ Interconnection Networks & Distributed Network Topology
An interconnection network's topology is the pattern by which its nodes, whether communicaโ€ฆ
Fault Tolerance & Dependable Computing at Scale
๐Ÿ›ก๏ธ Fault Tolerance & Dependable Computing at Scale
Fault tolerance is a computer system's ability to continue delivering correct service evenโ€ฆ
Parallelizing Compilers & Automatic Code Optimization
โš™๏ธ Parallelizing Compilers & Automatic Code Optimization
A parallelizing compiler analyses ordinary sequential source code to discover which of itsโ€ฆ
Grid/Wide-Area Distributed Resource Sharing
๐ŸŒ Grid/Wide-Area Distributed Resource Sharing
Grid computing pools computing, storage and instrument resources owned by many separate orโ€ฆ
Parallel & Distributed Storage Systems
๐Ÿ’ฝ Parallel & Distributed Storage Systems
A parallel or distributed storage system spreads a single logical file or dataset physicalโ€ฆ
Iterative Solvers & Large-Scale Matrix Computation
๐Ÿ”ข Iterative Solvers & Large-Scale Matrix Computation
An iterative solver finds an approximate solution to a large linear system of equations byโ€ฆ
Computer Architecture & Pipelined Processor Design for Scientific Computing
๐Ÿ–ฅ๏ธ Computer Architecture & Pipelined Processor Design for Scientific Computing
Instruction pipelining overlaps the several distinct stages needed to execute one instructโ€ฆ