Hello, How about?. I wanted to know if you could help me a problem. I have a machine with four graphics cards Tesla (no video output and Linux operating system) and another machine with a graphics card if you have video output (Linux operating system). Would connect the two machines so that the Tesla machine calculates the data and send to the second machine via Ethernet or some other type of network, so this shows the results in OpenGL. Do you know any way to do it hardware or software?
Sorry if I posted the message in the appropriate forum, I’m new yet.
hello avidday, thanks for answering so fast. The solution of nx server would serve in the event that the server machine, had graphics card with video output, but my case is reversed. I want to use the data to compute node Tesla (Tesla node has no video output) and the data sent to the client machine if you have video output.
I thought of sending a process CUDA by PVM to the host machine Tesla to perform calculations and send to the client machine and then display it using OpenGL. Could that be the solution?
hello avidday, thanks for answering so fast. The solution of nx server would serve in the event that the server machine, had graphics card with video output, but my case is reversed. I want to use the data to compute node Tesla (Tesla node has no video output) and the data sent to the client machine if you have video output.
I thought of sending a process CUDA by PVM to the host machine Tesla to perform calculations and send to the client machine and then display it using OpenGL. Could that be the solution?
I think you might have misunderstood how it works. The basic idea is that the compute machine runs as the nxserver and the local machine as the client. X11 (including GLX if you want) gets forwarded from the application on the compute server to be displayed on the local client with the video card. This is how we run out cluster - users run their compute and visualization jobs on headless cluster compute nodes with nx sending output to their local machines over ethernet. Isn’t that the model you want here?
I think you might have misunderstood how it works. The basic idea is that the compute machine runs as the nxserver and the local machine as the client. X11 (including GLX if you want) gets forwarded from the application on the compute server to be displayed on the local client with the video card. This is how we run out cluster - users run their compute and visualization jobs on headless cluster compute nodes with nx sending output to their local machines over ethernet. Isn’t that the model you want here?
I think you might have misunderstood how it works. The basic idea is that the compute machine runs as the nxserver and the local machine as the client. X11 (including GLX if you want) gets forwarded from the application on the compute server to be displayed on the local client with the video card. This is how we run out cluster - users run their compute and visualization jobs on headless cluster compute nodes with nx sending output to their local machines over ethernet. Isn’t that the model you want here?
I think VIrtualGL is the opposite of what you want. VirtualGL allows thin clients to use remote GPUs to render 3D and only forward the completed rendered frame to the client. That doesn’t sound like what you described in your first post to me.
I think VIrtualGL is the opposite of what you want. VirtualGL allows thin clients to use remote GPUs to render 3D and only forward the completed rendered frame to the client. That doesn’t sound like what you described in your first post to me.
Ok avidday, proves the MPI libraries to see if I can fix it. Anyway if you think of a better solution would be nice. Google will also continue looking at who knows everything XD