Search

In this study, we have introduced the Code-to-Code Adapter (CoToCoA) library to couple the magnetohydrodynamic (MHD) simulation and the Electron Hybrid (EH) simulation of planetary magnetospheres. CoToCoA has been developed newly to connect the different codes easily. The concept of CoToCoA is that we do not add modifications to each code as possible without data transfer functions, and we do not need to know the referred code without data format. With CoToCoA, we have been developing the cross-reference simulation of macro (MHD) and micro (EH) scales in the magnetosphere. Then, we have evaluated the performance of cross-reference simulation using CoToCoA on the massively parallel computer system.

We are now developing a manycore-aware implementation of multiprocessed PIC (particle-in-cell) simulation code with automatic load balancing. A key issue of the implementation is how to exploit the wide SIMD mechanism of manycore processors such as Intel Xeon Phi. Our solution is "particle binning" to rank all particles in a cell (voxel) in a chunk of SOA (structure-of-arrays) type one-dimensional arrays so that "particle-push" and "current-scatter" operations on them are efficiently SIMD-vectorized by our compiler. In addition, our sophisticated binning mechanism performs sorting of particles according to their positions "on-the-fly", efficiently coping with occasional "bin overflow" in a fully multithreaded manner. Our performance evaluation with up to 64 nodes of Cray XC30 and XC40 supercomputers, equipped with Xeon Phi 5120D (Knights Corner) and 7250 (Knights Landing) respectively, not only exhibited good parallel performance, but also proved the effectiveness of our binning mechanism.

We have been studying plasma flow response to a meso-scale magnetic dipole by means of full kinetic simulations using Particle-In-Cell method. The plasma flow response to a magnetic dipole and the resulting formation of a magnetosphere depends on the intensity of the magnetic moment of the dipole. The size of a magnetic dipole immersed in a plasma flow is characterized by a distance L from its center at which the equilibrium is satisfied between the pressure of the dipole magnetic field and that of the plasma flow. We particularly focused on a meso-scale magnetic dipole which implies that L is comparable to or somewhat smaller than the ion Larmor radius or the inertial length. In this case, plasma kinetics such as finite Larmor radius effect will play an important role to determine the plasma response to the magnetic dipole. Contrary to the case of the Earth's magnetosphere, we found that difference of dynamics between ions and electrons in the meso-scale dipole field plays an important role in the magnetosphere formation. In other words, electron-ion coupling through dipole fields becomes important. However, very little analysis has been done so far on the interactions between meso-scale dipole field and a plasma flow because plasma kinetics should be taken into account. To fully include the plasma kinetics in the analysis, we have been performing particle simulations for the investigation on the interactions between the solar wind and meso-scale dipole magnetic fields. The simulation results show that electron response to the local magnetic field is important in the process of meso-scale magnetosphere formation. Within the distance of L from the dipole center, charge separation occurs because of the difference of dynamics between magnetized electrons and unmagnetized ions in the meso-scale dipole fields. Then intense electrostatic field is induced inside the dipole region. It turns out that incoming ion flow to the dipole fields is eventually influenced by this intense electric field and the ions' trajectories are largely distorted. In two-dimensional simulations, we found the incoming ions are reflected by the electrostatic field at the distance L in the upstream region of the dipole. At the ion reflection point, magnetic fields are compressed, forming a magnetopause. The width of the boundary current layer as well as the spatial gradient of the local magnetic field compression found on the dayside magnetopause can be characterized by the electron Larmor radius and is independent of ion's spatial scale. We also examined the solar wind interactions with a magnetic anomaly called Reiner Gamma on the lunar surface. Since the magnetic field is almost perpendicular to the solar wind, increase of plasma and magnetic field densities is found at the dayside region. One of the interesting findings is that the solar wind ions hardly reach the moon surface in Reiner Gamma due to the interaction with the local field. Most ions are reflected back to the sunward direction at the magnetopause even though the ion Larmor radius is larger than L. We will discuss this point by considering the plasma dynamics as well as the electrostatic field observed over the Reiner Gamma region.

This paper describes a parallel implementation of our practical particle-in-cell (PIC) simulator with the OhHelp dynamic load-balancing algorithm. Although the code parallelization is based on simple block domain decomposition, OhHelp accomplishes load balancing and thus the scalability in terms of the number of particles by making each computation node help another heavily loaded node. In addition to the OhHelp application, a number of additional layers overlapping with adjacent domains are newly introduced outside the boundaries of each subdomain for the purpose of minimizing overhead costs of OhHelp. The optimization can drastically reduce overhead costs for particle transfer among nodes, whereas it leads to increase in domain size which each node is responsible for. Despite this trade-off feature, the overlapping layer attachment and a further lower-level optimization exert 1.8-fold improvement of the PIC simulator performance. Consequently, the optimized simulator exhibits a good scalability and a stable efficiency in parallel executions using up to 4096 cores, showing small parallel efficiency degradation of 3% from 16- to 4096-core parallel executions.

Recent expansion and sophistication of electric antenna applications in space missions increase the demand for better understanding of plasma and electromagnetic environment around the electric antenna and spacecraft. Because of strong plasma inhomogeneity emerging around the antenna and spacecraft, antenna behavior is difficult to resolve analytically and thus should be analyzed by means of some numerical approaches such as a plasma simulation. Actually, a practical merit of the plasma simulation has been demonstrated through its application to an analysis of antenna impedance in plasma [1]. Also, such a numerical approach has further advantages for an analysis of modern electric antennas equipped with distinctive devices: e.g., a photoelectron guard electrode, which greatly complicates the antenna-plasma interactions. In the present study, we apply the particle-in-cell simulations to the studies of the antenna and spacecraft environment. We particularly focus on electric properties of modern electric antennas used for static electric field measurements and plasma wave observations.