How to Improve Remote Visualization/Interactive Job Performance

At some point in your usage of the Nimbix cloud, you might decide that while running in batch mode is good, you may want to run your job in interactive mode. Running your job interactively allows you to see in real time what your model is doing, as well as enable you to interact remotely with your model. You might also be reading this article because you've launched an interactive job and the interface or model rotation appears to be slow and unusable. We'll walk you through some job setup tweaks which will improve your experience.


To achieve optimal job setup and performance when running a job interactively, like ANSYS Fluent, you'll want to select an NC* class of machine, like the nc5-us-01 selected below. The following screenshot shows you how to choose this type of machine during your job setup. We also call this type of machine "OpenGL on master."


By selecting this type of machine type, JARVICE is grabbing a GPU-enabled node and making that GPU-enabled node the master node of your job. This master-node handles all of the rendering server side for the job itself, no matter how many nodes/cores you've selected for your job. We find that using a GPU master-node will dramatically improve your job performance when it comes to interacting with the job, via a browser or VNC client.

To further enhance your interactive jobs, you might also consider using a VNC client, below is a quick article on how to do this.

Feel free to reach out to us at, should you have any additional questions!

Was this article helpful?
0 out of 0 found this helpful