Overview

Get Started

In order to speed up computation times, Flux offers its users the ability to parallelize and do some parametric distribution for heavy project computations.

Solver parallelization

During a finite element solving, several steps may be achieved in parallel (using several cores available on the machine) such as the building of the topological matrix, the assembly and the solving of the linear system, which usually is the most time-consuming task. All the steps described above directly depend on the number of nodes in the Flux project mesh. An optimal number of cores usually exists and allows to divide the computations among the different cores without spending too much time on the synchronization between them. This optimal number of cores reduces the calculation times as much as possible, without overloading the machine used.

This document uses 2D, 3D and Skew examples to give recommendations in terms of number of cores, linear solver and an estimation of the memory needed. To do this, a case has been selected for each application (Transient Magnetic, Steady-State AC magnetic and Magnetostatic). This case is solved several times with a different mesh, a different number of cores, from 1 to 32, and different solvers (direct or iterative).

Parametric Distribution on a single machine and on a cluster

The parametric distributed computing allows the user to save computation time while distributing several independent configurations of a same Flux project. For example a Magnetic Transient project may be distributed regarding different values of parameters such as the geometrical parameters (size, shape etc...) or physical parameters (supply current values, speed...) varying in the scenario. Several projects will run at the same time simulating all those configurations, the main parameter of a distributed computation is the number of running Flux in parallel.

This document shows the establishment of a parametric distribution on a single machine and on a cluster. For the single machine distribution, an internal tool of Flux is employed while PBS and other schedulers may be used on a cluster.