Optimization Simulation Performance in Simulink

Run Multiple Simulations at the Same Time

System design and testing often require engineers to run their Simulink models over multiple design configurations and operating conditions to improve and verify the design. Using MathWorks parallel computing products, you can exploit the repetitive nature of activities such as Monte Carlo simulations and design optimization studies by distributing work across multicore desktops and computer clusters.

Built-in parallel computing capabilities in many MathWorks products let you take advantage of parallel computing with little or no programming.

This paper was presented at the AIAA conference in August 2007. It discusses how to speed up executions of Simulink models. It describes and compares the use of Accelerator modes, RSim, and SystemTest with Distributed Computing Toolbox.

Explore Products for Parallel Computing

Generate Code in Parallel for Referenced Models

By using Parallel Computing Toolbox on a multicore desktop or MATLAB Parallel Server on a computer cluster, you can speed up code generation builds for Simulink models that contain large model reference hierarchies. You can use this capability to reduce diagram update times for simulation when the referenced models are in accelerated mode, and also for generating C/C++ code used for deployment to a DSP or microcontroller.