arXiv subjects

Yuejia Zhang

Publications and source records attributed to Yuejia Zhang.

3 recordsLinked to original sources

arXiv2026

CDFCI: High-Performance Parallel Software for Quantum Many-Body Eigenvalue Problems

CDFCI is a shared-memory parallel numerical program for computing low-lying eigenpairs of large-scale, non-relativistic fermionic Hamiltonians. The software is designed to handle a broad class of many-body quantum models, including both ab initio electronic structure Hamiltonians and lattice-based Hamiltonians arising in condensed matter physics. CDFCI combines an efficient coordinate-descent-based selected configuration interaction algorithm with dedicated parallelization strategies, achieving high performance on modern multi-core architectures. Benchmark results on representative quantum chemistry and condensed matter test cases demonstrate that CDFCI attains state-of-the-art accuracy with competitive performance compared to established selected configuration interaction (such as CIPSI or SHCI) and DMRG implementations. The software is open-source, extensively documented, and provides a Python interface for seamless integration with PySCF and other many-body simulation workflows.

physics.comp-ph↗

arXiv2025

Low-order reaction-diffusion system approximates heat transfer and flow structure in annular convection

Heat transfer in a fluid can be greatly enhanced by natural convection, giving rise to the nuanced relationship between the Nusselt number and Rayleigh number that has been a focus of modern fluid dynamics. Our work explores convection in an annular domain, where the geometry reinforces the large-scale circulatory flow pattern that is characteristic of natural convection. The flow must match the no-slip condition at the boundary, leading to a thin boundary layer where both the flow velocity and the temperature vary rapidly. To understand the system's heat transfer characteristics, we derive a reduced model from the Navier-Stokes-Boussinesq equations, whereby the equations of flow and heat are transformed to a system of low-order partial differential equations (PDEs) that take the form of a reaction-diffusion system. Solutions to the reaction-diffusion system, though they fail to predict dynamic events, preserve the same boundary-layer structure seen in the direct numerical simulation (DNS). By matching the solutions inside and outside the boundary layer, asymptotic analysis predicts a power-law relationship Nu $\propto$ Ra$^{1/4}$. Though difficult to distinguish from an exponent of 2/7, the predicted power law agrees well with measurements from DNS over several decades of the Rayleigh number. Considering the model's deficiencies in describing turbulent fluctuations and reversal events, the agreement regarding heat transfer characteristics is encouraging and suggests that the methodology of systematically deriving low-order PDEs from the governing equations may provide a useful complement to existing theories.

physics.flu-dyn↗

arXiv2025

Parallel Multi-Coordinate Descent Methods for Full Configuration Interaction

We develop a multi-threaded parallel coordinate descent full configuration interaction algorithm (mCDFCI), for the electronic structure ground-state calculation in the configuration interaction framework. The FCI problem is reformulated as an unconstrained minimization problem, and tackled by a modified block coordinate descent method with a deterministic compression strategy. mCDFCI is designed to prioritize determinants based on their importance, with block updates enabling efficient parallelization on shared-memory, multi-core computing infrastructure. We demonstrate the efficiency of the algorithm by computing an accurate benchmark energy for the chromium dimer in the Ahlrichs SV basis (48e, 42o), which explicitly includes $2.07 \times 10^9$ variational determinants. We also provide the binding curve of the nitrogen dimer under the cc-pVQZ basis set (14e, 110o). Benchmarks show up to $79.3\%$ parallel efficiency on 128 cores.

physics.chem-ph↗