High Performance Visualization

The high-performance visualization options let you enable parallel processing and choose where to postprocess the results data generated by your simulation. You view the results data in the Physics Results Explorer app.

This page discusses:

Execution

Important: For simulations run on the cloud from a structural simulation app, the results always run on the cloud server. For more information, see "About Location for Simulation Execution" in the Physics Simulation Fundamentals Guide.

Select the required option:

Run locally

Executes postprocessing on your computer.

By default, this option is selected.

Run remotely

Executes postprocessing on a different computer that has the Physics Results engine installed with a 3DOrchestrate Station.

Note: If you execute the simulation on your local computer, you cannot choose Run remotely to perform postprocessing on a different host computer.

Orchestration Services: The remote postprocessing is performed through 3DOrchestrate. Your system administrator must configure this network infrastructure with your on-premises 3DEXPERIENCE platform. There are two types of host computers that you can choose to process the results data:

Station name for Detached controller host
This type of computer must have the 3DEXPERIENCE platform native (rich client) Physics Simulation 3DEXPERIENCE platform installed. These 3DEXPERIENCE platform include Physics Results Explorer. The native apps installation includes an embedded (private) 3DOrchestrate Station as well as the Physics Results engine. By default, the station name is the same as the hostname of the computer. The 3DEXPERIENCE platform batch controller is executed on this computer.
Station name for Compute host
This type of computer must have two items installed: 1) a 3DOrchestrate Station, and 2) the Physics Results engine (which is installed from the media SIMULIA_PhysicsResults.) By default, the station name is the same as the hostname of the computer.

Use cloud resources for post-processing
Use a cloud server to perform results visualization processing. You can use a maximum of 2 cores on the cloud host for postprocessing.
Total number of cores
If you select Run locally, you can indicate the number of cores that you want to use to display your results. You can select the maximum number of cores available to you, up to a total of eight. However, this strategy can sometimes have an adverse effect on performance for other apps. In most cases, you can leave at least one core available for nonvisualization purposes. If the meshed model has fewer elements than the number of cores requested, it is processed using a single core.

You can specify the number of cores you want to use if you execute postprocessing on a remote host.

By default, this option has the value of 1.

Multi-core parameters

If the Total number of cores is set to 2 or more, the Multi-core parameters are available. Domain decomposition methods determine final load balancing on each of the nodes responsible for postprocessing operations. Select a Domain Decomposition method:

  • ParMETIS: Parallel library that implements a variety of algorithms for partitioning meshes. this option is selected.
  • RCB: Recursive Coordinate Bisection recursively halves the model using the longest dimension of the input model.

Enable field cache

If enabled, some animations may play faster as previously built animation data will be cached for future use. However, enabling field cache is memory intensive and performance may be compromised when working with larger models. This option is disabled by default.

Performance

Select the required option:

Use wireframe for view manipulation
Displays your results in wireframe style when performing view manipulations like zoom, pan, or rotate.
Hide labels for view manipulation
Hides view node and element labels when performing view manipulations like zoom, pan, or rotate.
Use decimated visualization for navigation
Drag the slider to apply more or less data compression to the rendition of your results displayed for navigation.
Use normal compression
Enable normal compression for postprocessing data. You can drag the slider to increase or decrease the compression factor. When selected, this option reduces the network bandwidth requirement, thereby increasing performance with a lower degree of visualization quality loss.
Use element order reduction
Enable element order reduction, which reduces the amount of postprocessing data transferred between the Physics Results engine host machine and client by reducing the number of triangles used to visualize second-order finite elements. When element order reduction is selected, contour plots will have less precision in high-gradient areas.
Use instancing

Enable a rendering technique that reduces the graphics memory required to render repetitive structures. If you enable instancing, you can also select Draw solid vector symbols. The solid arrows produced by this option, although easier to interpret, require more memory than rendering with wireframe arrows.

Enable rendering cache
Enable the rendering cache to use graphical data from memory to greatly improve performance. This option is selected by default. You can disable the cache to process larger models when memory resources are limited; however, performance declines.

Client Preferences

Automatic update tabular field exports

Generates automatic names for new meshes and groups under a mesh part and for new properties.

Automatically update all valid tabular field export objects after recomputing the simulation. Otherwise, when changes are made to exported field data, you must manually edit and update the Tabular Field Export objects from the tree or from the Feature Manager.

Display results activation warning message

Enables the display of a warning message when you are executing the postprocessing locally but are storing the results data on the 3DEXPERIENCE platform. The warning indicates the size of the download so that you can determine whether or not you want to proceed.