Expert Modelling: Hands-on help from Rocscience Engineers. Register Here
 

Search Results

GPU Acceleration

The GPU Acceleration feature is an experimental feature that uses your GPU to accelerate the speed of the field point solution when computing your model. This can potentially decrease the overall computation time of your analysis.

To turn on this feature:

Select: File > Preferences > Enable GPU Acceleration

GPU Acceleration

Requirements

This feature is only available for computers using a Nvidia graphics card and requires all drivers to be up to date.

This feature is most useful for analyses that use a large amount of field points. For example: when using a field point box.

NOTE:

This GPU Acceleration feature is an experimental feature that is not designed for all hardware, so please review the requirements noted above. If your hardware is not optimized, using this particular feature could cause the computation to be slower than it would be with the CPU, or it could cause the engine to crash during the computation process. This means that the computation will not be completed and the results of your analysis will not be shown. If this happens:

  1. Try updating your driver. Drivers must be up to date in order for the GPU Accelerator to function as expected.
  2. If updating the driver does not work:

  3. Turn off the feature by going to File > Preferences > Enable GPU Acceleration and toggle off the tool.
  4. You may then proceed with computing your saved model as per usual.

Performance Speeds using GPU Acceleration

The below table shows the improved speed of the field point solution for two samples models, one with a lower number of elements and one with a higher amount. Both models used 1 000 000 field points.

You’ll note that different Nvidia graphics cards produced different speed results for computing the field points solution. Specifically, higher performance graphics cards produced faster speeds.

For example, with the Nvidia Quadro K4000 card the speed of the field point solution was 4.5x faster using the GPU Acceleration feature, whereas with the GeForce RTX2080 the speed was 89x faster for the model with lower amount of elements and 70x faster for the model with the higher amount of elements.

Sample Performance Speeds for Field Point Solutions using GPU Acceleration

Model with 1207 elements

Model with 5942 elements

Baseline

CPU Performance

202.818s

879.684s

Nvidia Graphics Card

Quadro K4000

44.5s

178.733s

Quadro K1200

18.213s

49.512s

Quadro P2000

7.187s

26.038s

GeForce RTX2080

2.323s

12.614s

Account Icon - click here to log in or out of your account Shopping Cart icon Click here to search our site Click here to close Learning Tech Support Documentation Info Chevron Delete Back to Top View More" PDF File Calendar Location Language Fees Video Click here to visit Rocscience's LinkedIn page Click here to visit Rocscience's YouTube page Click here to visit Rocscience's Twitter page Click here to visit Rocscience's Facebook page Click here to visit Rocscience's Instagram page Bookmark Network Scroll down for more Checkmark Download Print Back to top Single User Multiple Users CPillar Dips EX3 RocFall RocPlane RocSupport RocTopple RS2 RS3 RSData RSPile Settle3 Slide2 Slide3 SWedge UnWedge Commercial License Education License Trial License Shop safe & secure Money-back guarantee