GPUQT: An efficient linear-scaling quantum transport code fully implemented on graphics processing units

Fan, Zheyong; Vierimaa, Ville; Harju, Ari

doi:10.1016/j.cpc.2018.04.013

Cited by 10 publications

(15 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pybinding [47], on the other hand, has a Python interface, and a C++ core, which besides basic diagonalization methods, can also use the kernel polynomial method to model finite size systems with disorder, strains or magnetic fields. GPUQT is a transport code fully implemented for the use on graphical processor units (GPUs), where the size of simulated domains are limited by the device memory to 2 × 10 7 [48]. ESSEX-GHOST [49] and TBTK [50] are C++ codes based on the kernel polynomial method that provide Chebyshev expansions of Green's functions for tight-binding models.…”

Section: Introductionmentioning

confidence: 99%

“…Pybinding [40], on the other hand, has a Python interface, and a C++ core, which besides basic diagonalization methods, can also use the kernel polynomial method to setup and model finite size systems with disorder, strains or magnetic fields. GPUQT is a transport code fully implemented for the use 3 on graphical processor units (GPUs), where the size of simulated domains are limited by the device memory to 2 × 10 7 [41]. and TBTK [43], are C++ codes based on the kernel polynomial method that provide Chebyshev expansions of the Green's functions for tight-binding models.Previous numerical implementations real-space quantum transport, either based on linear response theory or the non-equilibrium Green's function method [44], have so far been limited to mesoscopic structures with up to ten millions of orbitals [27,30,31,[45][46][47][48], hampering the extraction of maximum mileage from the tight-binding scheme (except for a recent report, where Kubo calculations with billions of atoms N = 3.6 × 10 9 were demonstrated [36]).…”

mentioning

confidence: 99%

See 1 more Smart Citation

KITE: high-performance accurate modelling of electronic structure and response functions of large molecules, disordered crystals and heterostructures

Anđelković

Covaci

et al. 2020

R. Soc. open sci.

View full text Add to dashboard Cite

We present KITE, a general purpose open-source tight-binding software for accurate real-space simulations of electronic structure and quantum transport properties of large-scale molecular and condensed systems with tens of billions of atomic orbitals (N ∼ 10 10 ). KITE's core is written in C++, with a versatile Python-based interface, and is fully optimised for shared memory multi-node CPU architectures, thus scalable, efficient and fast. At the core of KITE is a seamless spectral expansion of lattice Green's functions, which enables large-scale calculations of generic target functions with uniform convergence and fine control over energy resolution. Several functionalities are demonstrated, ranging from simulations of local density of states and photo-emission spectroscopy of disordered materials to large-scale computations of optical conductivity tensors and real-space wave-packet propagation in the presence of magneto-static fields and spin-orbit coupling. On-the-fly calculations of real-space Green's functions are carried out with an efficient domain decomposition technique, allowing KITE to achieve nearly ideal linear scaling in its multi-threading performance. Crystalline defects and disorder, including vacancies, adsorbates and charged impurity centers, can be easily set up with KITE's intuitive interface, paving the way to user-friendly large-scale quantum simulations of equilibrium and nonequilibrium properties of molecules, disordered crystals and heterostructures subject to a variety of perturbations and external conditions. arXiv:1910.05194v1 [cond-mat.mes-hall] 11 Oct 2019 2 IntroductionComputational modelling has become an essential tool in both fundamental and applied research that has propelled the discovery of new materials and their translation into practical applications [1]. The study of condensed phases of matter has benefited from significant advances in electronic structure theory and simulation methodologies. Among these advances are: explicitly correlated wave-function-based techniques achieving sub-chemical accuracy [2], first-principles methods to tackling electronic excitations [3], charge-self-consistent atomistic models for accurate electronic structure calculations [4], and the use of machine learning as means to finding density functionals without solving the Khon-Sham equations [5,6].Semi-empirical atomistic methods are amongst the most simple and effective methods to calculate ground-and excited-state properties of materials [7][8][9][10]. The increasingly popular tight-binding approach [11] has been employed for accurate and fast calculations of total energies and electronic structure in complex materials, including semiconductors [12,13], quantum dots [14] and super-lattices [15,16], and is particularly well-suited for implementation of O(N ) (linear scaling) algorithms for efficient calculations of total energies and forces [17].Accurate tight-binding models have been devised for a plethora of model systems, ranging from metals to ionic materials [18], and shown to correctly p...

show abstract

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

KITE: high-performance accurate modelling of electronic structure and response functions of large molecules, disordered crystals and heterostructures

Anđelković

Covaci

et al. 2020

R. Soc. open sci.

View full text Add to dashboard Cite

show abstract

“…The most important result is that the conductivity scales linearly with respect to the carrier density, giving a constant mobility. This transport fingerprint has been confirmed by using the LSQT method with the MSD formalism (Fan et al, 2017), as shown in Fig. 11.…”

Section: Charged Impuritiesmentioning

confidence: 58%

“…In addition, an implementation based on graphics processing units was used to obtain the results presented in this section and has been distributed as an open-source code named GPUQT (Fan et al, 2018).…”

Section: Implementationsmentioning

confidence: 99%

Linear scaling quantum transport methodologies

et al. 2021

Self Cite

View full text Add to dashboard Cite

“…By using (9), (6) and (12), (13) the density of states and electrical conductivity are estimated. To calculate the correlators [Û (t), X], the recursive algorithm is used (see [17,18]). Note that by limiting the delta function expansion with N m Chebyshev polynomials we achieve the finite energy resolution determined as dE ≈ π∆E/N m .…”

Section: Computational Formalismmentioning

confidence: 99%

Electron-hole asymmetry in electrical conductivity of low-fluorinated graphene: numerical study

Kolesnikov

Osipov

2020

Eur. Phys. J. B

View full text Add to dashboard Cite

By using the real-space Green-Kubo formalism we study numerically the electron transport properties of low-fluorinated graphene. At low temperatures the diffuse transport regime is expected to be prevalent, and we found a pronounced electron-hole asymmetry in electrical conductivity as a result of quasi-resonant scattering on the localized states. For the finite temperatures in the variable-range hopping transport regime the interpretation of numerical results leads to the appearance of local minima and maxima of the resistance near the energies of the localized states. A comparison with the experimental measurements of the resistance in graphene samples with various fluorination degrees is discussed.

show abstract

GPUQT: An efficient linear-scaling quantum transport code fully implemented on graphics processing units

Cited by 10 publications

References 32 publications

KITE: high-performance accurate modelling of electronic structure and response functions of large molecules, disordered crystals and heterostructures

KITE: high-performance accurate modelling of electronic structure and response functions of large molecules, disordered crystals and heterostructures

Linear scaling quantum transport methodologies

Electron-hole asymmetry in electrical conductivity of low-fluorinated graphene: numerical study

Contact Info

Product

Resources

About