A5084482336- Parallel Computing and Optimization Techniques
- Advanced Numerical Methods in Computational Mathematics
- Advanced Data Storage Technologies
- Distributed and Parallel Computing Systems
- Cardiac electrophysiology and arrhythmias
- Numerical methods for differential equations
- Interconnection Networks and Systems
- Matrix Theory and Algorithms
- Numerical methods in engineering
- Fractional Differential Equations Solutions
- Advanced Control Systems Optimization
- Computational Geometry and Mesh Generation
- Computational Physics and Python Applications
- Electromagnetic Simulation and Numerical Methods
- Cardiovascular Function and Risk Factors
- Differential Equations and Numerical Methods
- Embedded Systems Design Techniques
- Geological formations and processes
- Seismic Imaging and Inversion Techniques
- Electromagnetic Scattering and Analysis
- Advanced Condensed Matter Physics
- Advanced MRI Techniques and Applications
- VLSI and FPGA Design Techniques
- High-Velocity Impact and Material Behavior
- Physics of Superconductivity and Magnetism
Northern Jiangsu People's Hospital
2025
Simula Research Laboratory
2014-2024
University of Oslo
2013-2023
Nanchang Institute of Science & Technology
2019-2022
Xi'an High Tech University
2011-2015
Horological Research Institute of Light Industry
2012-2014
Zhejiang University
2013-2014
Beihang University
2014
Pennsylvania State University
2013
Xiamen University
2003-2007
We present Mint, a programming model that enables the non-expert to enjoy performance benefits of hand coded CUDA without becoming entangled in details. Mint targets stencil methods, which are an important class scientific applications. have implemented with source-to-source translator generates optimized C from traditional source. The relies on annotations guide translation at high level. set pragmas is small, and compact simple. Yet, able deliver competitive painstakingly hand-optimized...
Computational techniques have significantly advanced our understanding of cardiac electrophysiology, yet they predominantly concentrated on averaged models that do not represent the intricate dynamics near individual cardiomyocytes. Recently, accurate representing cells gained popularity, enabling analysis electrophysiology at micrometer level. Here, we evaluate five mathematical to determine their computational efficiency and physiological fidelity. Our findings reveal cell-based introduced...
This article addresses the performance of scientific applications that use Python programming language. First, we investigate several techniques for improving computational efficiency serial codes. Then, discuss basic in parallelizing applications. It is shown an efficient implementation array‐related operations essential achieving good parallel performance, as case. Once are efficiently implemented, probably using a mixed‐language implementation, and become achievable. confirmed by set...
The EMI model represents excitable cells in a more accurate manner than traditional homogenized models at the price of increased computational complexity. complexity solving stems from significant increase number nodes and form linear systems that need to be solved. Here, we will show latter problem can solved by careful use operator splitting spatially coupled equations. By using this method, broken into sub-problems are classical type linear, elliptic boundary value problems. Therefore,...
In this opinion paper we contend that high‐resolution characterization, monitoring, and prediction are the key elements to advancing reducing uncertainty in our understanding of subsurface processes at basin scales. First, advocate recently developed tomographic surveying is an effective approach for characterizing field‐scale subsurface. Fusion different types surveys further enhances characterization. A appropriate scale many water resources management purposes. We thereby propose...
The odd-parity, spin-triplet superconductor Sr${}_{2}$RuO${}_{4}$ has been found to feature exotic vortex physics including half-flux quanta trapped in a doubly connected sample and the formation of lattices at low fields. consequences these states on low-temperature magnetoresistive behavior mesoscopic samples were investigated by using ring device fabricated mechanically exfoliated single crystals photolithography focused ion beam. With magnetic field applied perpendicular in-plane...
Cardiomyocytes are the functional building blocks of heart-yet most models developed to simulate cardiac mechanics do not represent individual cells and their surrounding matrix. Instead, they work on a homogenized tissue level, assuming that cellular subcellular structures processes scale uniformly. Here we present mathematical numerical framework for exploring tissue-level microscale given an explicit three-dimensional geometrical representation embedded in We defined model over such...
Abstract The electrical activity in the heart is governed by bidomain equations. In this paper, we analyse an order optimal method for algebraic equations arising from discretization of model. Our scheme defined terms block Jacobi or symmetric Gauss–Seidel preconditioners. Furthermore, each these methods based on standard preconditioners scalar elliptic parabolic partial differential (PDEs). Such can be realized multigrid domain decomposition schemes, and are thus readily available applying...
In the context of multiple GPUs that share same PCIe bus, we propose a new communication scheme leads to more effective overlap and computation. Multiple CUDA streams OpenMP threads are adopted so data can simultaneously be sent received. A representative 3D stencil example is used demonstrate effectiveness our scheme. We compare performance with an MPI-based state-of-the-art Results show approach outperforms scheme, being up 1.85× faster. However, results also indicate current underlying...
A recent trend in modern high-performance computing environments is the introduction of powerful, energy-efficient hardware accelerators such as GPUs and Xeon Phi coprocessors. These specialized devices coexist with CPUs are optimized for highly parallel applications. In regular computing-intensive applications predictable data access patterns, these often far outperform thus relegate latter to pure control functions instead computations. For irregular applications, however, performance gap...