The electron-phonon Wannier interpolation (EPWI) method is an efficient way to compute electron-phonon interaction (EPI) properties accurately. In this study, we present a GPU-accelerated implementation of the EPWI method for computing transport properties, followed by a performance analysis. The implementation is tested on common systems such as aluminum and silicon. The results show complete consistency with those obtained through CPU computations. The proposed algorithm computes the conductivity of aluminum in 20 min on a single NVIDIA Tesla V100 GPU, using a ${200}^3$ electron and phonon sampling grid. This is 54 times faster than the CPU-based algorithm running on two nodes of Intel Xeon Platinum 8260 CPUs. The impressive acceleration is achieved by carefully designing the algorithm to reduce the computational workload and exploit the specific features of the GPU. The approach can also be extended to GPU codes for other EPI-related properties.

Close

PrevNext

View figure in article

• Electron-phonon coupling
• First-principles calculations
• Metals
• Semiconductors
• Boltzmann theory
• Density functional calculations