On the 7th of October we launched a MATLAB® toolbox for simulating models with OnScale. To learn more about the toolbox you can read our first introductory blog or keep reading to find out the top features of the toolbox.
Customization and flexibility
One of the main features of the toolbox is the ability for users to customize to suit their workflow. The toolbox was designed by engineers for engineers and provides all the features needed for general simulations. However, this can be expanded on by users, who can create their own functions to fit their own needs, using the existing code as inspiration.
The flexibility provided by scripting is unmatched and allows users to get exactly what they want out of their simulations. MATLAB® is a widely used programming environment with a huge catalog of functions. Many of these functions are very useful when it comes to Finite Element Analysis (FEA) simulations and analysis. MATLAB® also have a wide range of toolboxes which are powerful tools for design and analysis such as the Optimization ToolboxTM and the Statistics and Machine Learning ToolboxTM which can be easily integrated with the OnScale toolbox.
The MATLAB® environment also allows for easy and seamless pre and post processing. This is a huge benefit especially for the set-up of huge optimization problems such as Monte Carlo studies.
Start to finish workflow
Another beneficial feature of the toolbox is the full start to finish workflow. Users can create a model, submit it to the cloud, download the results and post-process them in one script. This drastically speeds up the time to go through the full workflow and minimizes the need for user input.
Users can create their own optimization routines to run simulations, analyze results and re-run until desired results are found. Data can also be kept extremely organized by setting up custom automatic save locations and there is no need to transfer data between different programs.
Ease of use
The toolbox has simplified many aspects of the simulation process such as submitting, downloading and post processing jobs. These processes can be executed in just a couple of lines of code. Submitting jobs using Message Passing Interface (MPI) is also now a simplified process, as well as the creation of hybrid mesh models.
OnScale is a CAE software that makes use of scalable high-performance computing (HPC) resources to aid in the innovation and design of next-generation technologies. The limitless power of the cloud allows engineers to run 100s and 1000s of simulations to optimize designs and find solutions quickly. Along with this comes the need to be able to post-process 100s and 1000s of results. The toolbox can automatically process all simulations in a job in one line of code without the need for nested loops.
Another feature of the toolbox is the ability to add keypoints to a model automatically. Keypoints are an important part of modelling in OnScale. They are points that define both the XYZ co-ordinates and IJK nodal indices of important locations in the model such as:
- Important geometry considerations
- Specific points for data extraction
- Specific regions for different meshing considerations
It is specifically important to map keypoints to important geometry considerations as this allows for easy parameterization and mesh control.
Adding keypoints to a model can be time consuming for complex models but the toolbox has the option to automatically detect and add keypoints to user models.
The final exciting feature of the toolbox is the ability to slice 3D models. Jumping straight in to modelling in 3D can be difficult as there are so many considerations. We always recommend simulating in 2D first and ensure that the results are as expected before moving onto more complex geometry. It is also a lot quicker to simulate in 2D so this is ideal for working through any design uncertainties or issues that might occur. If you use the toolbox for modelling you only need to create one 3D model and then slice it to get a 2D model. This allows you to easily switch between 2D and 3D without the need to create and maintain to separate model files.
These are just a few of the top features of the MATLAB® Toolbox for Cloud Engineering Simulation with OnScale. You can get your hands on the toolbox now for free! Just head over to our Getting Started documentation to download and install.