Parallelization of a local gyrokinetic Vlasov simulation code for Peta-scale computing

Maeyama, Shinya; Watanabe, Tomohiko*; Idomura, Yasuhiro   ; Nakata, Motoki; Nunami, Masanori*; Ishizawa, Akihiro*

To address multi-scale turbulence including temporal and spatial scales of electrons and ions, we extend our gyrokinetic Vlasov simulation code GKV to run efficiently on Peta-scale supercomputers. A key numerical technique is the parallel two-dimensional Fast Fourier Transform (FFT) required for evaluating the nonliear EB advection term, where masking of the cost of inter-node transpose communications is essential to improve strong scaling. To mask communication costs, overlaps of computations and transpose communications is implemented with the help of the hybrid parallelization of message passing interface and open multiprocessing. Integrated overlaps including FFTs and calculations of the nonlinear EB term show a better scaling than simple overlaps of transpose and FFTs. The masking of communication costs significantly improves strong scaling of the GKV code, and makes substantial speed-up toward multi-scale turbulence simulations.