Public Software
Particle-Based Remote Visualization Applications: CS/IS-PBVR
CS/IS PBVR is a visualization application based on Particle-Based Volume Rendering (PBVR). It is possible to visualize the calculation result data (volume data) of large-scale 3D simulations at high speed and 3D point cloud data acquired by laser measurements. You can choose between Client Server (CS) visualization, which visualizes volume data stored in remote storage, and In-Situ (IS), which visualizes simulations simultaneously and in the same environment. It also supports VR visualization with a head-mounted display.
Linear equations library for large sparse problems : PARCEL
The PARallel Computing ELements (PARCEL) library provides highly efficient parallel Krylov subspace solvers for modern massively parallel supercomputers, which are characterized by accelerated computation and less performance improvement in inter-node communication. The PARCEL is based on a hybrid parallel programing model with MPI+OpenMP, and supports the latest communication avoiding (CA) Krylov methods (the Chebyshev basis CACG method and the CA-GMRES method) in addition to the conventional Krylov subspace methods (the CG method, the BiCGstab method, and the GMRES method). The PARCEL supports two matrix formats (Compressed Row Storage (CRS) formant and Diagonal (DIA) format), two data types (Double precision and Quad precision), and can be called from programs written in C and Fortran.
※ The latest version is 1.2
Quadruple Precision BLAS Routines (QPBLAS)
A quadruple precision version of the Basic Linear Algebra Subroutines (BLAS). It is expected that, when a simulation that requires the K computer or very-large-scale parallel computers of the future to run at their maximum performance is conducted, the accuracy of the calculation can be significantly reduced in some cases. It is necessary to improve accuracy of calculation in this situation and quadruple precision with higher speed of basic linear calculation can be achieved by using this software.
※ The latest version is 1.0
Quadruple Precision BLAS Routines for GPU (QPBLAS-GPU)
A quadruple precision version of the Basic Linear Algebra Subroutines (BLAS) on GPUs. By simply replacing the conventional BLAS routines with corresponding QPBLAS-GPU routines, quadruple precision with higher speed can be achieved. In addition, not only basic operations but also applications that support GPUs can utilize quadruple precision calculations, and users can easily conduct quadruple precision calculations on GPUs.
※ The latest version is 1.0
Parallel Molecular Simulations : PIMD
This is a general software for ab initio molecular simulation, published an open source code written in MPI Fortran 90 . A variety of simulation methods is implemented, such as geometry optimization, normal mode analysis, phonon calculations, reaction path calculations (string method, path integral method), classical statistics (replica exchange), quantum statistics (path integral method) semiclassical dynamics (centroid and ring polymer molecular dynamics), nonadiabatic dynamics (mean field dynamics method, surface hopping method), free energy calculations (metadynamics, constrained molecular dynamics), etc. Each of them employs algorithms suited for massively parallel computers.
※ The latest version is 2.6.0